query
stringclasses 259
values | prequel
listlengths 0
30
| sequel
listlengths 0
30
| during
listlengths 0
30
| metadata
dict | num_of_frames
int64 4
4
| frame_captions
listlengths 4
4
| video_id
stringlengths 6
45
| frame2video_caption
stringlengths 157
1.26k
|
|---|---|---|---|---|---|---|---|---|
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.
|
[
"People running out of buildings during the 2016 Gyeongju, South Korea earthquake on September 12",
"Emergency responders searching through rubble after the 2016 Gyeongju, South Korea earthquake on September 12, 2016",
"Cars driving away from the affected area in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Buildings with cracked walls and broken windows in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Rescue teams carrying injured people on stretchers during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"Buildings collapsing or showing significant structural damage during the 12 September 2016 earthquake in Gyeongju, South Korea",
"People evacuating their homes and gathering in safe areas in Gyeongju, South Korea after the earthquake on September 12, 2016",
"Emergency responders searching for survivors in rubble after the 2016 Gyeongju, South Korea earthquake on September 12",
"Roads blocked by debris and damaged vehicles after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Residents assessing damage to their homes and belongings after the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"People running out of buildings during the earthquake in Gyeongju, South Korea on September 12, 2016",
"Rescue teams searching through rubble after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Buildings with cracked walls and fallen debris in Gyeongju, South Korea after the 12 September 2016 earthquake",
"Emergency vehicles rushing to the scene after the 2016 Gyeongju, South Korea earthquake on September 12",
"People gathering in open spaces for safety during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the consequences and aftermath of an earthquake, as these are the most likely to be visualized in a video. Events such as damage to buildings, rescue operations, and people reacting to the quake are plausible and commonly seen in earthquake-related videos.\n\nEVENTS:\n1. People running out of buildings\n2. Rescue teams searching through rubble\n3. Buildings with cracked walls and fallen debris\n4. Emergency vehicles rushing to the scene\n5. People gathering in open spaces for safety",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the aftermath or impact of an earthquake, as these would be the most visually apparent and relevant to a video about the 2016 Gyeongju earthquake. Events such as destruction, evacuation, and rescue efforts are likely to be seen in a video associated with this query.\n\nEVENTS:\n1. People running out of buildings\n2. Emergency responders searching through rubble\n3. Cars driving away from the affected area\n4. Buildings with cracked walls and broken windows\n5. Rescue teams carrying injured people on stretchers",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the aftermath and impact of an earthquake, such as damage to infrastructure, rescue efforts, and evacuation procedures. These events are likely to be visually represented in a video associated with the query, providing a concrete and tangible depiction of the earthquake's effects.\n\nEVENTS:\n1. Buildings collapsing or showing significant structural damage\n2. People evacuating their homes and gathering in safe areas\n3. Emergency responders searching for survivors in rubble\n4. Roads blocked by debris and damaged vehicles\n5. Residents assessing damage to their homes and belongings",
"video_URL": "http://youtube.com/watch?v=t8lY-rjy7-Y",
"video_description": "9월 12일, 경주에서는 역대 최대 규모인 5.8 지진이 일어났다. 지진으로 서울과 수도권까지 전국이 흔들렸고, 이로 인해 전화와 카카오톡도 모두 먹통이었다. 정부는 지진 발생 9분이 지나서야 긴급재난문자를 발송했고 이마저도 서울과 수도권 지역 국민들은 문자를 받지 못했다. #경주 #지진 #재난문자",
"video_id": "t8lY-rjy7-Y"
}
| 4
|
[
"The image shows a surveillance camera view of an aisle in a warehouse or storage area. The shelves on the right are stocked with numerous boxes, and the area appears organized. On the left, there is a wall with some equipment and boxes. The text at the bottom of the image indicates the time and location in Korean, translating to \"7:44 PM, Ganghwa City, Namnam-dong, 9km away from the scene.\"",
"The image is a black and white surveillance camera view inside a room. A person is seen in the center of the frame, kneeling on a bed. The person appears to be interacting with a small object on the bed, which could be a toy or an item of interest. The room is relatively dimly lit, with the primary light source coming from the upper right corner of the image. There are a few objects on the bed, including what looks like a stuffed animal and some pillows. The text at the bottom of the image is in Korean, translating to \"Seoul and Su-dogun are also being searched.\" The overall setting suggests a domestic environment, possibly a bedroom, and the person seems to be engaged in a quiet or focused activity.",
"The image shows a person sitting at a desk, speaking into a microphone, and holding a pen. The setting appears to be an official or professional environment, indicated by the presence of a laptop on the desk and a flag in the background. The person is wearing a white blouse and seems to be engaged in delivering a speech or making an official statement. The text at the bottom of the image is in Korean, translating to \"Even in the larger space, you are the most qualified one in front of the camera.\" The scene suggests that the person is addressing an audience or participating in a formal event.",
"The person continues to sit at the desk, still speaking into the microphone and holding a pen, indicating they are delivering a speech or making an official statement. The setting remains consistent with the previous frame, maintaining the official and professional environment with the laptop and flag in the background. The individual remains engaged in addressing the audience or participating in the formal event."
] |
t8lY-rjy7-Y
|
The video appears to depict a series of unrelated scenes. It begins with a surveillance view of a warehouse or storage area, showing an organized aisle with shelves and equipment. The scene then shifts to a domestic environment, possibly a bedroom, where a person is kneeling on a bed, interacting with a small object in a quiet and focused manner. The video then cuts to an official or professional setting, where a person is sitting at a desk, delivering a speech or making an official statement into a microphone. This person continues to address the audience in the final frame, maintaining a professional demeanor in front of a flag and laptop. The overall context suggests a narrative that jumps between different locations and activities, possibly indicating a search or investigation in progress, as hinted by the Korean text translations.
|
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.
|
[
"People running out of buildings during the 2016 Gyeongju, South Korea earthquake on September 12",
"Emergency responders searching through rubble after the 2016 Gyeongju, South Korea earthquake on September 12, 2016",
"Cars driving away from the affected area in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Buildings with cracked walls and broken windows in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Rescue teams carrying injured people on stretchers during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"Buildings collapsing or showing significant structural damage during the 12 September 2016 earthquake in Gyeongju, South Korea",
"People evacuating their homes and gathering in safe areas in Gyeongju, South Korea after the earthquake on September 12, 2016",
"Emergency responders searching for survivors in rubble after the 2016 Gyeongju, South Korea earthquake on September 12",
"Roads blocked by debris and damaged vehicles after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Residents assessing damage to their homes and belongings after the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"People running out of buildings during the earthquake in Gyeongju, South Korea on September 12, 2016",
"Rescue teams searching through rubble after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Buildings with cracked walls and fallen debris in Gyeongju, South Korea after the 12 September 2016 earthquake",
"Emergency vehicles rushing to the scene after the 2016 Gyeongju, South Korea earthquake on September 12",
"People gathering in open spaces for safety during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the consequences and aftermath of an earthquake, as these are the most likely to be visualized in a video. Events such as damage to buildings, rescue operations, and people reacting to the quake are plausible and commonly seen in earthquake-related videos.\n\nEVENTS:\n1. People running out of buildings\n2. Rescue teams searching through rubble\n3. Buildings with cracked walls and fallen debris\n4. Emergency vehicles rushing to the scene\n5. People gathering in open spaces for safety",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the aftermath or impact of an earthquake, as these would be the most visually apparent and relevant to a video about the 2016 Gyeongju earthquake. Events such as destruction, evacuation, and rescue efforts are likely to be seen in a video associated with this query.\n\nEVENTS:\n1. People running out of buildings\n2. Emergency responders searching through rubble\n3. Cars driving away from the affected area\n4. Buildings with cracked walls and broken windows\n5. Rescue teams carrying injured people on stretchers",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the aftermath and impact of an earthquake, such as damage to infrastructure, rescue efforts, and evacuation procedures. These events are likely to be visually represented in a video associated with the query, providing a concrete and tangible depiction of the earthquake's effects.\n\nEVENTS:\n1. Buildings collapsing or showing significant structural damage\n2. People evacuating their homes and gathering in safe areas\n3. Emergency responders searching for survivors in rubble\n4. Roads blocked by debris and damaged vehicles\n5. Residents assessing damage to their homes and belongings",
"video_URL": "http://youtube.com/watch?v=CXe6_CDrdB8",
"video_description": "찐♡기상청사람들 예고 없이 찾아오는 지진! 한 번 발생하면 큰 피해를 주는 지진이기에 감시하는 일은 한시도 소홀히 할 수 없습니다. 그 최전방에서... 1년 365일 24시간 지진화산 감시 및 분석 업무에 최선을 다해 신속·정확하게 알리고자 노력하는 사람이 있습니다. 지진화산 감시 및 분석 업무 9년 차... 반복되는 주간 근무와 야간 근무가 힘들긴 하지만, 누구보다 묵묵하게 자리를 지키고 있는 진득한 뚝배기 같은 지진업무의 대들보!! 지진화산 감시 업무를 담당하는 기상청 사람을 만나고 왔습니다.(^_^)v 우리나라에서 가장 강력했던 규모 5.8의 경주 지진 때 지진감시 및 분석을 수행했던 기상청 사람은 누구? 지진감시 업무를 들어올 때마다 긴장되는 기상청 사람은 누구? 국민의 안전을 누구보다 걱정하는 기상청 사람은 누구? 지진 올바로 알고 지진 발생 시 빠르게 대처하길 바라는 기상청 사람은 누구? 최근 지진 안전을 홍보해 주시는 지진희 배우님을 그 누구보다 고마워하는 기상청 사람은 누구? 만나볼까요 :) #기상청사람들 #지진화산 #행동하는파수꾼",
"video_id": "CXe6_CDrdB8"
}
| 4
|
[
"The image shows a scene of a busy street or market area, with several people walking. There is text in Korean that appears to discuss record times or distances. The text mentions \"5.8 km\" and \"5.8 hours,\" suggesting a comparison or achievement in a specific context, possibly related to travel or distance running.",
"In the current frame, a man wearing a blue shirt is standing in front of a computer screen that displays various charts and graphs. The text on the screen is in Korean and discusses the need to take a break from work, suggesting that the man might be in a work environment, possibly dealing with data analysis or monitoring. The man is smiling, indicating a positive or light-hearted moment.",
"The man in a blue shirt is still standing in front of the computer screen with charts and graphs. He now appears more serious, looking directly at the camera. The Korean text on the screen mentions taking a break from work, possibly indicating he is discussing work-life balance or emphasizing the importance of downtime. His expression suggests he is conveying a message or making a point.",
"The current frame features a close-up of handwritten Korean text on a spiral notebook. The text seems to be a continuation of the discussion about work-life balance, specifically focusing on the concept of a \"ginsang champion.\" It mentions the idea of leaving work earlier and dealing with more work during limited times. The text encourages the reader to be more efficient in their tasks and to seek ways to handle the demands of their job more effectively. The hashtags at the bottom suggest that this is a post from social media or a blog, possibly aimed at inspiring or informing viewers about productivity and time management."
] |
CXe6_CDrdB8
|
A man discusses work-life balance and productivity in a series of scenes, starting with a busy street or market area displaying a record time or distance achievement. He then appears in a work environment, standing in front of a computer screen showing charts and graphs, initially smiling and later looking serious as he conveys a message about taking breaks from work. The video also includes a close-up of handwritten text on a notebook, where the concept of a "ginsang champion" is introduced, emphasizing the importance of efficiency, time management, and handling work demands effectively. The overall context suggests the man is sharing insights or advice on how to balance work and personal life, possibly in a social media or blog post.
|
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.
|
[
"People running out of buildings during the 2016 Gyeongju, South Korea earthquake on September 12",
"Emergency responders searching through rubble after the 2016 Gyeongju, South Korea earthquake on September 12, 2016",
"Cars driving away from the affected area in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Buildings with cracked walls and broken windows in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Rescue teams carrying injured people on stretchers during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"Buildings collapsing or showing significant structural damage during the 12 September 2016 earthquake in Gyeongju, South Korea",
"People evacuating their homes and gathering in safe areas in Gyeongju, South Korea after the earthquake on September 12, 2016",
"Emergency responders searching for survivors in rubble after the 2016 Gyeongju, South Korea earthquake on September 12",
"Roads blocked by debris and damaged vehicles after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Residents assessing damage to their homes and belongings after the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"People running out of buildings during the earthquake in Gyeongju, South Korea on September 12, 2016",
"Rescue teams searching through rubble after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Buildings with cracked walls and fallen debris in Gyeongju, South Korea after the 12 September 2016 earthquake",
"Emergency vehicles rushing to the scene after the 2016 Gyeongju, South Korea earthquake on September 12",
"People gathering in open spaces for safety during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the consequences and aftermath of an earthquake, as these are the most likely to be visualized in a video. Events such as damage to buildings, rescue operations, and people reacting to the quake are plausible and commonly seen in earthquake-related videos.\n\nEVENTS:\n1. People running out of buildings\n2. Rescue teams searching through rubble\n3. Buildings with cracked walls and fallen debris\n4. Emergency vehicles rushing to the scene\n5. People gathering in open spaces for safety",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the aftermath or impact of an earthquake, as these would be the most visually apparent and relevant to a video about the 2016 Gyeongju earthquake. Events such as destruction, evacuation, and rescue efforts are likely to be seen in a video associated with this query.\n\nEVENTS:\n1. People running out of buildings\n2. Emergency responders searching through rubble\n3. Cars driving away from the affected area\n4. Buildings with cracked walls and broken windows\n5. Rescue teams carrying injured people on stretchers",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the aftermath and impact of an earthquake, such as damage to infrastructure, rescue efforts, and evacuation procedures. These events are likely to be visually represented in a video associated with the query, providing a concrete and tangible depiction of the earthquake's effects.\n\nEVENTS:\n1. Buildings collapsing or showing significant structural damage\n2. People evacuating their homes and gathering in safe areas\n3. Emergency responders searching for survivors in rubble\n4. Roads blocked by debris and damaged vehicles\n5. Residents assessing damage to their homes and belongings",
"video_URL": "http://youtube.com/watch?v=uF3x63vtddc",
"video_description": "\"경주 지진, 지하수는 1~2일 전에 미리 알았다\" 지난 12일과 19일 경주에서 규모 5.8을 비롯한 대형 지진이 잇따라 발생하기 하루 이틀 전 주변 지하수 수위가 크게 상승해, 향후 지진예측에 활용할 수 있다는 주장이 나왔습니다. 정상용 부경대 지구환경과학과 교수는 진앙에서 가장 가까운 경북 경주시 산내면에 있는 지하수 관측소 자료를 분석한 결과, 이 같이 밝혔습니다. 자료를 보면 지난 11일 암반 지하수 수위는 하루 사이에 무려 49㎝ 상승했고, 17일에도 39㎝나 올라갔습니다. 정 교수는 지진에 앞서 암석에 강한 압력이 가해져 지하수 수위가 많이 올라간 것으로 보인다며, 규모 4.5 이상의 지진에만 영향을 받았다고 설명했습니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "uF3x63vtddc"
}
| 4
|
[
"The image shows a night vision scene depicting a small area with a structure that resembles a shrine or altar. There is a bright light or fire near the structure, possibly indicating activity or ritual. The image also includes text in Korean which translates to \"Jeongju crime scene, suspect arrived one to two days ago.\"",
"The image shows an interior setting featuring a decorative mirror with intricate designs, a portion of a white air conditioning unit, and a window with blinds. The scene appears to be part of an investigative report, indicated by the Korean text on the screen which translates to \"Jeongju crime scene, suspect arrived one to two days ago.\" This suggests a connection to a crime investigation, with the current frame focusing on a specific location within the area of interest.",
"The image shows a street scene with a white car parked on the side of the road. Several individuals are gathered around the vehicle, some of whom appear to be engaged in conversation or inspection. The setting is urban, with storefronts visible in the background. Korean text on the screen translates to \"Jeongju crime scene, suspect arrived one to two days ago.\" This indicates that the location is part of a crime scene investigation, and the individuals may be involved in or observing the investigation process.",
"The image continues to show the street scene with the white car parked on the side of the road. Several individuals are still gathered around the vehicle, suggesting ongoing activity. The urban setting with storefronts remains visible in the background. The Korean text reiterates \"Jeongju crime scene, suspect arrived one to two days ago,\" reinforcing the context of a crime scene investigation. The individuals present might be investigators, witnesses, or onlookers connected to the ongoing investigation."
] |
uF3x63vtddc
|
A crime investigation is underway in Jeongju, where a suspect arrived one to two days ago. The scene begins with a nighttime setting, showing a structure resembling a shrine or altar with a fire or bright light, indicating possible ritualistic activity. The investigation then shifts to an interior location, focusing on a decorative mirror and window, suggesting a search for evidence or clues. The scene moves to a street setting, where a white car is parked and several individuals, potentially investigators or witnesses, are gathered around the vehicle, engaged in conversation or inspection. The ongoing activity and presence of multiple individuals imply a thorough investigation into the crime, with the suspect's recent arrival in Jeongju being a key factor in the case.
|
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.
|
[
"People running out of buildings during the 2016 Gyeongju, South Korea earthquake on September 12",
"Emergency responders searching through rubble after the 2016 Gyeongju, South Korea earthquake on September 12, 2016",
"Cars driving away from the affected area in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Buildings with cracked walls and broken windows in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Rescue teams carrying injured people on stretchers during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"Buildings collapsing or showing significant structural damage during the 12 September 2016 earthquake in Gyeongju, South Korea",
"People evacuating their homes and gathering in safe areas in Gyeongju, South Korea after the earthquake on September 12, 2016",
"Emergency responders searching for survivors in rubble after the 2016 Gyeongju, South Korea earthquake on September 12",
"Roads blocked by debris and damaged vehicles after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Residents assessing damage to their homes and belongings after the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"People running out of buildings during the earthquake in Gyeongju, South Korea on September 12, 2016",
"Rescue teams searching through rubble after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Buildings with cracked walls and fallen debris in Gyeongju, South Korea after the 12 September 2016 earthquake",
"Emergency vehicles rushing to the scene after the 2016 Gyeongju, South Korea earthquake on September 12",
"People gathering in open spaces for safety during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the consequences and aftermath of an earthquake, as these are the most likely to be visualized in a video. Events such as damage to buildings, rescue operations, and people reacting to the quake are plausible and commonly seen in earthquake-related videos.\n\nEVENTS:\n1. People running out of buildings\n2. Rescue teams searching through rubble\n3. Buildings with cracked walls and fallen debris\n4. Emergency vehicles rushing to the scene\n5. People gathering in open spaces for safety",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the aftermath or impact of an earthquake, as these would be the most visually apparent and relevant to a video about the 2016 Gyeongju earthquake. Events such as destruction, evacuation, and rescue efforts are likely to be seen in a video associated with this query.\n\nEVENTS:\n1. People running out of buildings\n2. Emergency responders searching through rubble\n3. Cars driving away from the affected area\n4. Buildings with cracked walls and broken windows\n5. Rescue teams carrying injured people on stretchers",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the aftermath and impact of an earthquake, such as damage to infrastructure, rescue efforts, and evacuation procedures. These events are likely to be visually represented in a video associated with the query, providing a concrete and tangible depiction of the earthquake's effects.\n\nEVENTS:\n1. Buildings collapsing or showing significant structural damage\n2. People evacuating their homes and gathering in safe areas\n3. Emergency responders searching for survivors in rubble\n4. Roads blocked by debris and damaged vehicles\n5. Residents assessing damage to their homes and belongings",
"video_URL": "http://youtube.com/watch?v=nln3QFJggKs",
"video_description": "106억 → 58억…경주 지진 피해금액 반토막 잇단 강진으로 100억원이 넘는 것으로 추정됐던 경주시의 공식 피해규모가 58억원 수준에 그치는 것으로 집계됐습니다. 경주시가 지진 발생 뒤 전수조사를 통해 발표한 잠정 추산 피해액은 약 107억원이었지만 사유재산 피해는 대부분 공식 피해액에서 빠졌습니다. 규정상 기와가 떨어지거나 벽체에 금이 가고 담장이 파손된 것은 재산피해 산정되지 않습니다. 이에 따라 사유재산 중에는 지진으로 주택이 완전히 부서지거나 반쯤 허물어진 것만 공식 피해액에 들어갔습니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "nln3QFJggKs"
}
| 4
|
[
"The image shows a construction scene where a crane is lifting a large object over a building with a traditional Korean-style roof. Several vehicles, including trucks, are parked in the foreground, and some people are visible in the vicinity. The text in the top left corner reads \"경주 4.5여진 안전처 홈피 또 먹통\" indicating an incident related to a 4.5 magnitude earthquake in Gyeongju. The bottom text mentions a damage cost of 106 billion won being revised to 58 billion won, with the cause attributed to a lack of public bidding.",
"The current frame depicts an outdoor scene in front of a traditional Korean house. A man is seen leaning against the wall, appearing to be in the middle of a conversation with an older individual seated on a chair. A woman stands nearby, holding what seems to be a tool or a piece of wood. The setting appears to be a rural area with a stack of wood or branches piled on the right side. The background includes lush greenery and another building with a tiled roof, suggesting a village or countryside environment. The text in the top left corner reads \"경주 4.5여진 안전처 홈피 또 먹통,\" indicating an ongoing issue related to a 4.5 magnitude earthquake in Gyeongju. The bottom text still mentions the damage cost being revised from 106 billion won to 58 billion won due to a lack of public bidding.",
"The current frame shows a close-up view of a cracked road surface, highlighting the aftermath of the 4.5 magnitude earthquake in Gyeongju. The visible text reinforces the ongoing issue, mentioning the damage cost being revised from 106 billion won to 58 billion won due to a lack of public bidding. The context suggests the impact of the earthquake on infrastructure, emphasizing the need for rebuilding and public safety measures.",
"The current frame shows a close-up view of a wall with tiles that appear to be damaged or displaced, likely due to the earthquake. The visible text indicates that there has been significant damage caused by the 4.5 magnitude earthquake, which has affected cultural assets and infrastructure. The context continues to emphasize the need for rebuilding and public safety measures in the aftermath of the earthquake."
] |
nln3QFJggKs
|
The video depicts the aftermath of a 4.5 magnitude earthquake in Gyeongju, showcasing the damage to infrastructure and cultural assets. The scene shifts between a construction site where a crane is lifting debris, a rural area with traditional Korean houses, and close-up views of cracked roads and damaged walls. The earthquake has caused significant destruction, with an initial damage cost estimate of 106 billion won being revised to 58 billion won due to a lack of public bidding. The video highlights the need for rebuilding and public safety measures, emphasizing the impact of the earthquake on the community and the importance of addressing the damage to ensure public safety.
|
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.
|
[
"People running out of buildings during the 2016 Gyeongju, South Korea earthquake on September 12",
"Emergency responders searching through rubble after the 2016 Gyeongju, South Korea earthquake on September 12, 2016",
"Cars driving away from the affected area in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Buildings with cracked walls and broken windows in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Rescue teams carrying injured people on stretchers during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"Buildings collapsing or showing significant structural damage during the 12 September 2016 earthquake in Gyeongju, South Korea",
"People evacuating their homes and gathering in safe areas in Gyeongju, South Korea after the earthquake on September 12, 2016",
"Emergency responders searching for survivors in rubble after the 2016 Gyeongju, South Korea earthquake on September 12",
"Roads blocked by debris and damaged vehicles after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Residents assessing damage to their homes and belongings after the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"People running out of buildings during the earthquake in Gyeongju, South Korea on September 12, 2016",
"Rescue teams searching through rubble after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Buildings with cracked walls and fallen debris in Gyeongju, South Korea after the 12 September 2016 earthquake",
"Emergency vehicles rushing to the scene after the 2016 Gyeongju, South Korea earthquake on September 12",
"People gathering in open spaces for safety during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the consequences and aftermath of an earthquake, as these are the most likely to be visualized in a video. Events such as damage to buildings, rescue operations, and people reacting to the quake are plausible and commonly seen in earthquake-related videos.\n\nEVENTS:\n1. People running out of buildings\n2. Rescue teams searching through rubble\n3. Buildings with cracked walls and fallen debris\n4. Emergency vehicles rushing to the scene\n5. People gathering in open spaces for safety",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the aftermath or impact of an earthquake, as these would be the most visually apparent and relevant to a video about the 2016 Gyeongju earthquake. Events such as destruction, evacuation, and rescue efforts are likely to be seen in a video associated with this query.\n\nEVENTS:\n1. People running out of buildings\n2. Emergency responders searching through rubble\n3. Cars driving away from the affected area\n4. Buildings with cracked walls and broken windows\n5. Rescue teams carrying injured people on stretchers",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the aftermath and impact of an earthquake, such as damage to infrastructure, rescue efforts, and evacuation procedures. These events are likely to be visually represented in a video associated with the query, providing a concrete and tangible depiction of the earthquake's effects.\n\nEVENTS:\n1. Buildings collapsing or showing significant structural damage\n2. People evacuating their homes and gathering in safe areas\n3. Emergency responders searching for survivors in rubble\n4. Roads blocked by debris and damaged vehicles\n5. Residents assessing damage to their homes and belongings",
"video_URL": "http://youtube.com/watch?v=n6UmQ9Kw2GQ",
"video_description": "경주 지진 이후 여진 307회 발생…\"강도 약해져\" 지난 12일 저녁 발생한 규모 5.8의 경주 지진 탓에 여진이 307차례 발생한 것으로 집계됐습니다. 기상청에 따르면 오늘 오후 4시까지 경주 지진 이후 여진은 규모 1.5∼3.0도 292회 등 모두 307차례 발생했습니다. 기상청 관계자는 \"경주 지진 탓에 여진이 계속 발생하고 있지만 강도가 약해지고 있다\"며 \"앞으로 일어날 여진 규모는 3.0 이하로 예상되기 때문에 그렇게 우려하지 않아도 될 것 같다\"고 말했습니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "n6UmQ9Kw2GQ"
}
| 4
|
[
"The image shows a kitchen setting with various kitchen utensils and appliances visible. There is a cutting board, utensils hanging on the wall, and pots and pans on the counter. A news ticker is present at the bottom of the image with Korean text. The top right corner has additional text in Korean, likely indicating a news broadcast or television program.",
"The current frame shows a ceiling light fixture with a white lampshade featuring a floral pattern, hanging from a ceiling bracket. The setting appears to be indoors, possibly in a residential or office environment. There is visible text on the screen in Korean, indicating a news broadcast or television program, with additional details on the top right corner suggesting a specific news segment or time.",
"The current frame shows an indoor setting with a potted plant near a wall-mounted device. There is a bright light source coming from a screen or monitor on the right side. The scene appears to be part of a news broadcast or television program, indicated by the visible Korean text and a news ticker at the bottom of the screen. The text and ticker provide information in Korean, suggesting the news is related to a specific event or incident.",
"The video is part of a news broadcast on 연합뉴스TV (Yonhap News TV) focusing on a significant event or incident involving a woman who gave birth to 307 children after a sexual assault. The scene shows an indoor setting with a potted plant and a bright light source from a screen, likely to be part of the news studio setup. The text and ticker provide further information about the event, highlighting the distressing nature of the incident."
] |
n6UmQ9Kw2GQ
|
A news broadcast on Yonhap News TV reports on a disturbing incident involving a woman who gave birth to 307 children after a sexual assault, with the news studio setup showing an indoor environment with various props and a screen displaying relevant information in Korean.
|
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.
|
[
"People running out of buildings during the 2016 Gyeongju, South Korea earthquake on September 12",
"Emergency responders searching through rubble after the 2016 Gyeongju, South Korea earthquake on September 12, 2016",
"Cars driving away from the affected area in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Buildings with cracked walls and broken windows in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Rescue teams carrying injured people on stretchers during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"Buildings collapsing or showing significant structural damage during the 12 September 2016 earthquake in Gyeongju, South Korea",
"People evacuating their homes and gathering in safe areas in Gyeongju, South Korea after the earthquake on September 12, 2016",
"Emergency responders searching for survivors in rubble after the 2016 Gyeongju, South Korea earthquake on September 12",
"Roads blocked by debris and damaged vehicles after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Residents assessing damage to their homes and belongings after the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"People running out of buildings during the earthquake in Gyeongju, South Korea on September 12, 2016",
"Rescue teams searching through rubble after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Buildings with cracked walls and fallen debris in Gyeongju, South Korea after the 12 September 2016 earthquake",
"Emergency vehicles rushing to the scene after the 2016 Gyeongju, South Korea earthquake on September 12",
"People gathering in open spaces for safety during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the consequences and aftermath of an earthquake, as these are the most likely to be visualized in a video. Events such as damage to buildings, rescue operations, and people reacting to the quake are plausible and commonly seen in earthquake-related videos.\n\nEVENTS:\n1. People running out of buildings\n2. Rescue teams searching through rubble\n3. Buildings with cracked walls and fallen debris\n4. Emergency vehicles rushing to the scene\n5. People gathering in open spaces for safety",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the aftermath or impact of an earthquake, as these would be the most visually apparent and relevant to a video about the 2016 Gyeongju earthquake. Events such as destruction, evacuation, and rescue efforts are likely to be seen in a video associated with this query.\n\nEVENTS:\n1. People running out of buildings\n2. Emergency responders searching through rubble\n3. Cars driving away from the affected area\n4. Buildings with cracked walls and broken windows\n5. Rescue teams carrying injured people on stretchers",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the aftermath and impact of an earthquake, such as damage to infrastructure, rescue efforts, and evacuation procedures. These events are likely to be visually represented in a video associated with the query, providing a concrete and tangible depiction of the earthquake's effects.\n\nEVENTS:\n1. Buildings collapsing or showing significant structural damage\n2. People evacuating their homes and gathering in safe areas\n3. Emergency responders searching for survivors in rubble\n4. Roads blocked by debris and damaged vehicles\n5. Residents assessing damage to their homes and belongings",
"video_URL": "http://youtube.com/watch?v=uoagwCVoHGM",
"video_description": "경주 지진으로 8명 부상…재산피해 253건 어제 저녁 경주 일대에서 발생한 지진으로 인해 8명이 다치고 카카오톡과 국민안전처 홈페이지 등 통신망이 일시적으로 마비되는 등의 장애가 있었습니다. 국민안전처는 경주 일대에서 오늘과 어제에 걸쳐 일어난 지진으로 인해 80살 여성 황 모 씨 등 8명이 경상을 입고 253건의 재산피해가 신고됐다고 밝혔습니다. 이번 지진으로 인해 KTX 대전~부산 구간이 5분가량 정차하기도 했습니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "uoagwCVoHGM"
}
| 4
|
[
"The frame shows a grainy black and white image of a chaotic scene, likely a news report or footage from a disaster area. The text in Korean at the bottom of the screen reads \"경주 지진으로 8명 부상...재산피해 253건\", which translates to \"8 people injured due to Gyeongju earthquake...253 property damage cases\". The image appears to depict debris and damage, possibly indicating the aftermath of an earthquake.",
"The current frame depicts a person using a broom to clean up debris on a tiled floor. The individual is wearing bright blue shoes and appears to be actively sweeping the area. The context suggests that the person is engaged in cleanup efforts following the earthquake mentioned in the previous frame. The Korean text at the bottom of the screen remains consistent with the previous frame, reinforcing the ongoing theme of the aftermath of the Gyeongju earthquake and the associated damages.",
"The current frame shows a close-up of a cracked wall, highlighting the damage caused by the Gyeongju earthquake. The text at the bottom of the screen continues to mention the earthquake's aftermath, specifically noting that eight buildings have been demolished and 253 have been repaired. The visual focus on the structural damage underscores the extent of the earthquake's impact on buildings in the region.",
"In this frame, three individuals are seen walking side by side through a hallway. The person in the middle has their arm around the individual on their right, suggesting a supportive or friendly gesture. The setting appears to be an indoor environment, possibly within a residential or office building. The individuals are dressed casually, and there is a sense of normalcy in their demeanor, despite the ongoing discussion of a significant event, as indicated by the text at the bottom of the screen. This text likely pertains to a news broadcast, given the presence of a timestamp and other informational elements."
] |
uoagwCVoHGM
|
The video depicts the aftermath of an earthquake in Gyeongju, showing the damage and cleanup efforts. The footage includes a chaotic scene with debris, a person sweeping up debris, and a close-up of a cracked wall, highlighting the extent of the destruction. The earthquake resulted in 8 people injured and 253 property damage cases. The video also shows individuals walking together in a hallway, suggesting a sense of normalcy and support amidst the disaster. Overall, the video appears to be a news report or documentary about the Gyeongju earthquake, showcasing the impact on the community and the ongoing recovery efforts.
|
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.
|
[
"People running out of buildings during the 2016 Gyeongju, South Korea earthquake on September 12",
"Emergency responders searching through rubble after the 2016 Gyeongju, South Korea earthquake on September 12, 2016",
"Cars driving away from the affected area in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Buildings with cracked walls and broken windows in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Rescue teams carrying injured people on stretchers during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"Buildings collapsing or showing significant structural damage during the 12 September 2016 earthquake in Gyeongju, South Korea",
"People evacuating their homes and gathering in safe areas in Gyeongju, South Korea after the earthquake on September 12, 2016",
"Emergency responders searching for survivors in rubble after the 2016 Gyeongju, South Korea earthquake on September 12",
"Roads blocked by debris and damaged vehicles after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Residents assessing damage to their homes and belongings after the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"People running out of buildings during the earthquake in Gyeongju, South Korea on September 12, 2016",
"Rescue teams searching through rubble after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Buildings with cracked walls and fallen debris in Gyeongju, South Korea after the 12 September 2016 earthquake",
"Emergency vehicles rushing to the scene after the 2016 Gyeongju, South Korea earthquake on September 12",
"People gathering in open spaces for safety during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the consequences and aftermath of an earthquake, as these are the most likely to be visualized in a video. Events such as damage to buildings, rescue operations, and people reacting to the quake are plausible and commonly seen in earthquake-related videos.\n\nEVENTS:\n1. People running out of buildings\n2. Rescue teams searching through rubble\n3. Buildings with cracked walls and fallen debris\n4. Emergency vehicles rushing to the scene\n5. People gathering in open spaces for safety",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the aftermath or impact of an earthquake, as these would be the most visually apparent and relevant to a video about the 2016 Gyeongju earthquake. Events such as destruction, evacuation, and rescue efforts are likely to be seen in a video associated with this query.\n\nEVENTS:\n1. People running out of buildings\n2. Emergency responders searching through rubble\n3. Cars driving away from the affected area\n4. Buildings with cracked walls and broken windows\n5. Rescue teams carrying injured people on stretchers",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the aftermath and impact of an earthquake, such as damage to infrastructure, rescue efforts, and evacuation procedures. These events are likely to be visually represented in a video associated with the query, providing a concrete and tangible depiction of the earthquake's effects.\n\nEVENTS:\n1. Buildings collapsing or showing significant structural damage\n2. People evacuating their homes and gathering in safe areas\n3. Emergency responders searching for survivors in rubble\n4. Roads blocked by debris and damaged vehicles\n5. Residents assessing damage to their homes and belongings",
"video_URL": "http://youtube.com/watch?v=5p1N4MtudiI",
"video_description": "경주 지진 316회 이어가다 현재는 '잠잠' 지난 12일 발생한 규모 5.8의 경주 지진의 여진이 316회 발생한 것으로 집계됐습니다. 기상청은 오늘 오전 11시 기준 여진 횟수가 316회라고 발표했습니다. 규모별 횟수는 1.5에서 3.0이 301차례, 3.0에서 4.0은 14차례, 4.0에서 5.0은 1차례 감지됐습니다. 사흘째 계속된 여진은 일단 멈추는 징후를 보이고 있는 것으로 기상청은 파악하고 있습니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "5p1N4MtudiI"
}
| 4
|
[
"The image shows a news broadcast screen displaying a radar map and seismic activity data. On the left, there's a map with a network of lines and data points indicating a seismic event. The right side features a detailed radar map with concentric circles and indicators, suggesting tracking of an object or phenomenon. Text overlays at the bottom and top provide information about the news event, mentioning seismic activity and the current temperature.",
"The image displays a news broadcast screen focusing on seismic activity data. The screen is divided into sections with various seismographs showing waveforms and annotations like \"Rg,\" \"Pb,\" and \"Lg,\" indicating different types of seismic waves. The text overlay in Korean reads \"정주 지진 316회 이어가다 현재는 잠잠\" which translates to \"The Zhongju earthquake continues for the 316th time, currently quiet.\" This suggests that the video is discussing an ongoing series of earthquakes in Zhongju, noting a temporary lull in activity.",
"The current frame shows a map of a geographical area marked with numerous yellow and orange circles, likely representing seismic activity or earthquake epicenters. The text overlay in Korean reads \"정주 지진 316회 이어가다 현재는 잠잠\" which translates to \"The Zhongju earthquake continues for the 316th time, currently quiet.\" This suggests the video is providing a visual representation of the widespread seismic activity, indicating the continued occurrence of earthquakes in Zhongju with a current period of inactivity.",
"The map continues to show the geographical area with numerous yellow and orange circles, indicating ongoing seismic activity or earthquake epicenters. The text overlay in Korean reads \"정주 지진 316회 이어가다 현재는 잠잠\" which translates to \"The Zhongju earthquake continues for the 316th time, currently quiet.\" This reinforces the video's focus on the widespread seismic activity in Zhongju, highlighting the continuation of earthquakes with a current period of inactivity."
] |
5p1N4MtudiI
|
A news broadcast reports on an ongoing series of earthquakes in Zhongju, noting that the seismic activity has continued for the 316th time with a current period of relative quiet. The video displays seismic activity data, including radar maps and seismographs, to provide visual context to the event. Maps of the affected geographical area are shown, marked with numerous yellow and orange circles representing earthquake epicenters, to illustrate the widespread nature of the seismic activity.
|
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.
|
[
"People running out of buildings during the 2016 Gyeongju, South Korea earthquake on September 12",
"Emergency responders searching through rubble after the 2016 Gyeongju, South Korea earthquake on September 12, 2016",
"Cars driving away from the affected area in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Buildings with cracked walls and broken windows in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Rescue teams carrying injured people on stretchers during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"Buildings collapsing or showing significant structural damage during the 12 September 2016 earthquake in Gyeongju, South Korea",
"People evacuating their homes and gathering in safe areas in Gyeongju, South Korea after the earthquake on September 12, 2016",
"Emergency responders searching for survivors in rubble after the 2016 Gyeongju, South Korea earthquake on September 12",
"Roads blocked by debris and damaged vehicles after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Residents assessing damage to their homes and belongings after the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"People running out of buildings during the earthquake in Gyeongju, South Korea on September 12, 2016",
"Rescue teams searching through rubble after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Buildings with cracked walls and fallen debris in Gyeongju, South Korea after the 12 September 2016 earthquake",
"Emergency vehicles rushing to the scene after the 2016 Gyeongju, South Korea earthquake on September 12",
"People gathering in open spaces for safety during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the consequences and aftermath of an earthquake, as these are the most likely to be visualized in a video. Events such as damage to buildings, rescue operations, and people reacting to the quake are plausible and commonly seen in earthquake-related videos.\n\nEVENTS:\n1. People running out of buildings\n2. Rescue teams searching through rubble\n3. Buildings with cracked walls and fallen debris\n4. Emergency vehicles rushing to the scene\n5. People gathering in open spaces for safety",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the aftermath or impact of an earthquake, as these would be the most visually apparent and relevant to a video about the 2016 Gyeongju earthquake. Events such as destruction, evacuation, and rescue efforts are likely to be seen in a video associated with this query.\n\nEVENTS:\n1. People running out of buildings\n2. Emergency responders searching through rubble\n3. Cars driving away from the affected area\n4. Buildings with cracked walls and broken windows\n5. Rescue teams carrying injured people on stretchers",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the aftermath and impact of an earthquake, such as damage to infrastructure, rescue efforts, and evacuation procedures. These events are likely to be visually represented in a video associated with the query, providing a concrete and tangible depiction of the earthquake's effects.\n\nEVENTS:\n1. Buildings collapsing or showing significant structural damage\n2. People evacuating their homes and gathering in safe areas\n3. Emergency responders searching for survivors in rubble\n4. Roads blocked by debris and damaged vehicles\n5. Residents assessing damage to their homes and belongings",
"video_URL": "http://youtube.com/watch?v=roaxxb8zRLU",
"video_description": "국제신문 뉴스&이슈 구독하기 http://goo.gl/FFQL4 국제신문 근교산&기획특집 구독하기 http://goo.gl/SM0fM 국제신문 기사보기 http://iii.im/pk3Z 국제신문 영상팀이 경주 지진피해 복구 현장을 드론으로 담았다.",
"video_id": "roaxxb8zRLU"
}
| 4
|
[
"An aerial view of a traditional tiled roof with some sections covered by blue tarpaulin, indicating possible repairs or maintenance. A white car is parked on the street adjacent to the building.",
"The current frame shows several workers on a traditional tiled roof covered with blue tarpaulin, engaged in repair or maintenance work. The workers are wearing safety helmets and vests, indicating a focus on safety during their task. The text in the image suggests the theme is about \"disaster prevention,\" indicating the work might be related to preparing for or recovering from a disaster.",
"The current frame shows an aerial view of a rural area with a road running through fields of green crops and a few buildings. There is a large building with a blue roof on the left and a multi-story school building on the right. A few buses are parked near the school building, and a roundabout is visible near the intersection. The text in the image suggests the theme is about \"Kangju Jinamji,\" indicating a focus on the location and possibly the disaster prevention measures taken there.",
"The aerial view continues to show the rural area with the road running through fields of green crops. The building with the blue roof remains on the left, while the multi-story school building is still visible on the right, with buses parked nearby. The roundabout near the intersection is also present. The text \"Kangju Jinamji\" and \"Namyangju Namyangju High School - Flood prevention, infrastructure improvement\" emphasizes the focus on disaster prevention and infrastructure enhancements in this area."
] |
roaxxb8zRLU
|
The video depicts a rural area, specifically Kangju Jinamji, where disaster prevention and infrastructure improvement efforts are underway. The scene opens with a close-up view of a traditional tiled roof undergoing repairs or maintenance, with workers prioritizing safety. The camera then pans out to reveal the broader context: a rural landscape with a road, fields, and notable buildings such as a school. The focus is on preparations for or recovery from disasters, particularly flooding, as indicated by the text and the presence of buses and infrastructure near the school. The video highlights the importance of disaster prevention measures and infrastructure enhancements in the area, showcasing the community's efforts to mitigate the impact of natural disasters.
|
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.
|
[
"People running out of buildings during the 2016 Gyeongju, South Korea earthquake on September 12",
"Emergency responders searching through rubble after the 2016 Gyeongju, South Korea earthquake on September 12, 2016",
"Cars driving away from the affected area in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Buildings with cracked walls and broken windows in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Rescue teams carrying injured people on stretchers during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"Buildings collapsing or showing significant structural damage during the 12 September 2016 earthquake in Gyeongju, South Korea",
"People evacuating their homes and gathering in safe areas in Gyeongju, South Korea after the earthquake on September 12, 2016",
"Emergency responders searching for survivors in rubble after the 2016 Gyeongju, South Korea earthquake on September 12",
"Roads blocked by debris and damaged vehicles after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Residents assessing damage to their homes and belongings after the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"People running out of buildings during the earthquake in Gyeongju, South Korea on September 12, 2016",
"Rescue teams searching through rubble after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Buildings with cracked walls and fallen debris in Gyeongju, South Korea after the 12 September 2016 earthquake",
"Emergency vehicles rushing to the scene after the 2016 Gyeongju, South Korea earthquake on September 12",
"People gathering in open spaces for safety during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the consequences and aftermath of an earthquake, as these are the most likely to be visualized in a video. Events such as damage to buildings, rescue operations, and people reacting to the quake are plausible and commonly seen in earthquake-related videos.\n\nEVENTS:\n1. People running out of buildings\n2. Rescue teams searching through rubble\n3. Buildings with cracked walls and fallen debris\n4. Emergency vehicles rushing to the scene\n5. People gathering in open spaces for safety",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the aftermath or impact of an earthquake, as these would be the most visually apparent and relevant to a video about the 2016 Gyeongju earthquake. Events such as destruction, evacuation, and rescue efforts are likely to be seen in a video associated with this query.\n\nEVENTS:\n1. People running out of buildings\n2. Emergency responders searching through rubble\n3. Cars driving away from the affected area\n4. Buildings with cracked walls and broken windows\n5. Rescue teams carrying injured people on stretchers",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the aftermath and impact of an earthquake, such as damage to infrastructure, rescue efforts, and evacuation procedures. These events are likely to be visually represented in a video associated with the query, providing a concrete and tangible depiction of the earthquake's effects.\n\nEVENTS:\n1. Buildings collapsing or showing significant structural damage\n2. People evacuating their homes and gathering in safe areas\n3. Emergency responders searching for survivors in rubble\n4. Roads blocked by debris and damaged vehicles\n5. Residents assessing damage to their homes and belongings",
"video_URL": "http://youtube.com/watch?v=3ojHTwrvx90",
"video_description": "박 대통령, 경주 지진현장 방문 \"특별재난지역 선포\" 박근혜 대통령은 최근 강진이 잇따라 발생한 경북 경주 지역을 찾아 특별재난지역 선포를 적극 검토하라고 지시했습니다. 이어 지진 등 재난 안전 시스템을 원점에서 재점검할 것도 주문했습니다. 박 대통령은 아울러 인근 월성 원자력발전소를 찾아 \"원전은 단 한치의 실수도 용납되지 않는다\"며 지진 방재대책을 철저히 재점검해달라고 당부했습니다. 박 대통령의 지시에 국민안전처 또한 경주시를 특별재난지역으로 선포할 예정입니다. 특별재난지역으로 지정되면 지자체의 피해 복구비 부담이 완화되고, 피해 주민들은 공공요금 감면혜택을 받게됩니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "3ojHTwrvx90"
}
| 4
|
[
"The image shows a group of people walking together outdoors. They are all wearing yellow jackets. The scene appears to be a public or official event, possibly a visit or inspection. A banner and some text in Korean at the bottom of the image suggest a formal context, mentioning \"President Park Geun-hye, Gyeongju City Hall visit 'Declaration of Special Zone for the World'\". The setting seems to be a street with buildings and vehicles in the background.",
"The image depicts a group of people, most of whom are wearing yellow jackets, gathered closely together. The central focus is on a woman who appears to be addressing the group or an audience, as indicated by her posture and the attention of those around her. The text overlay in Korean reads \"President Park Geun-hye, Gyeongju City Hall visit 'Declaration of Special Zone for the World',\" indicating that the event is related to a formal announcement or ceremony. The presence of microphones and the formal attire of the participants suggest a significant public or governmental event.",
"The group of people, most of whom are wearing yellow jackets, continue to be gathered closely together. They appear to be listening attentively, possibly to the woman who had been addressing them. The text overlay in Korean reads \"President Park Geun-hye, Gyeongju City Hall visit 'Declaration of Special Zone for the World',\" indicating that the event is related to a formal announcement or ceremony. The presence of microphones and the formal attire of the participants suggest a significant public or governmental event.",
"The group of people, most of whom are wearing yellow jackets, are now gathered around a man who appears to be the focus of their attention. The man is holding a microphone, indicating that he is addressing the group. The text overlay in Korean reads \"President Park Geun-hye, Gyeongju City Hall visit 'Declaration of Special Zone for the World',\" indicating that the event is related to a formal announcement or ceremony. The presence of microphones and the formal attire of the participants suggest a significant public or governmental event."
] |
3ojHTwrvx90
|
The video depicts a formal event where President Park Geun-hye visits Gyeongju City Hall to declare a special zone for the world. A group of people, mostly wearing yellow jackets, gather to participate in or witness the event. The scene involves a series of addresses or speeches, first by a woman and then by a man, both using microphones to convey their messages to the audience. The event appears to be a significant public or governmental ceremony, with participants showing keen interest and attention to the declarations being made. The overall context suggests an official visit and announcement related to designating a special zone, with the involvement of high-ranking officials and a formal setting.
|
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.
|
[
"People running out of buildings during the 2016 Gyeongju, South Korea earthquake on September 12",
"Emergency responders searching through rubble after the 2016 Gyeongju, South Korea earthquake on September 12, 2016",
"Cars driving away from the affected area in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Buildings with cracked walls and broken windows in Gyeongju, South Korea after the earthquake on 12 September 2016",
"Rescue teams carrying injured people on stretchers during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"Buildings collapsing or showing significant structural damage during the 12 September 2016 earthquake in Gyeongju, South Korea",
"People evacuating their homes and gathering in safe areas in Gyeongju, South Korea after the earthquake on September 12, 2016",
"Emergency responders searching for survivors in rubble after the 2016 Gyeongju, South Korea earthquake on September 12",
"Roads blocked by debris and damaged vehicles after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Residents assessing damage to their homes and belongings after the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
[
"People running out of buildings during the earthquake in Gyeongju, South Korea on September 12, 2016",
"Rescue teams searching through rubble after the 12 September 2016 earthquake in Gyeongju, South Korea",
"Buildings with cracked walls and fallen debris in Gyeongju, South Korea after the 12 September 2016 earthquake",
"Emergency vehicles rushing to the scene after the 2016 Gyeongju, South Korea earthquake on September 12",
"People gathering in open spaces for safety during the 12 September 2016 earthquake in Gyeongju, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the consequences and aftermath of an earthquake, as these are the most likely to be visualized in a video. Events such as damage to buildings, rescue operations, and people reacting to the quake are plausible and commonly seen in earthquake-related videos.\n\nEVENTS:\n1. People running out of buildings\n2. Rescue teams searching through rubble\n3. Buildings with cracked walls and fallen debris\n4. Emergency vehicles rushing to the scene\n5. People gathering in open spaces for safety",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the aftermath or impact of an earthquake, as these would be the most visually apparent and relevant to a video about the 2016 Gyeongju earthquake. Events such as destruction, evacuation, and rescue efforts are likely to be seen in a video associated with this query.\n\nEVENTS:\n1. People running out of buildings\n2. Emergency responders searching through rubble\n3. Cars driving away from the affected area\n4. Buildings with cracked walls and broken windows\n5. Rescue teams carrying injured people on stretchers",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2016 Gyeongju earthquake, it makes sense to extract events that are directly related to the aftermath and impact of an earthquake, such as damage to infrastructure, rescue efforts, and evacuation procedures. These events are likely to be visually represented in a video associated with the query, providing a concrete and tangible depiction of the earthquake's effects.\n\nEVENTS:\n1. Buildings collapsing or showing significant structural damage\n2. People evacuating their homes and gathering in safe areas\n3. Emergency responders searching for survivors in rubble\n4. Roads blocked by debris and damaged vehicles\n5. Residents assessing damage to their homes and belongings",
"video_URL": "http://youtube.com/watch?v=w6jdwGYcKZU",
"video_description": "안전처 \"경주지진 부상자 23명ㆍ재산피해 1,118건\" 국민안전처는 경주에서 발생한 규모 5.8의 지진으로 부상자 23명이 발생하고, 재산상 피해는 1천118건으로 집계됐다고 밝혔습니다. 안전처와 보건복지부에 따르면 부상자 23명 가운데 7명은 병원에서 치료를 받고 귀가했으며, 16명은 입원 중입니다. 지역별 부상자는 경북이 12명으로 가장 많고, 울산 4명, 대구 2명, 부산·인천·충북·전남·경남 각 1명입니다. 안전처는 균열이 발생한 주택을 대상으로 구조기술사 등과 함께 안전점검을 하고 귀가 여부와 추가 안전조치 등을 검토할 예정입니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "w6jdwGYcKZU"
}
| 4
|
[
"The image appears to show a scene with debris and rubble, indicating possible destruction or a disaster aftermath. There are fragments of structures and scattered objects on the ground. The text at the bottom of the image, written in Korean, states \"경주 지진 부상자 23명 · 재산피해 1,118건\" which translates to \"Gyeongju earthquake, 23 injured, 1,118 property damages.\" Additionally, there is a news ticker at the top providing information about a high-speed road accident involving a bus.",
"The current frame depicts a close-up view of a surface covered with what appears to be snow or frost. The ground is uniformly white, suggesting cold weather conditions. There are no visible structures or debris, indicating a different environment from the previous frame which showed destruction. The text at the bottom of the image, written in Korean, states \"경주 지진 부상자 23명 · 재산피해 1,118건\" which translates to \"Gyeongju earthquake, 23 injured, 1,118 property damages.\" This suggests that the current frame is part of a news segment about the aftermath of the Gyeongju earthquake.",
"The current frame shows a wall with a noticeable crack, indicating structural damage. A hand is pointing at the crack, highlighting the extent of the damage. The text at the bottom of the image, written in Korean, reiterates \"경주 지진 부상자 23명 · 재산피해 1,118건\" which translates to \"Gyeongju earthquake, 23 injured, 1,118 property damages.\" This frame continues the narrative of the news segment, focusing on the structural impact of the earthquake.",
"The current frame depicts a scene of significant destruction following an earthquake. A gray SUV is parked on a gravel surface, surrounded by debris and broken roof tiles scattered across the ground, suggesting extensive damage to nearby structures. The backdrop includes traditional Korean-style houses with some visible damage, reinforcing the impact of the earthquake. The text at the bottom of the image, written in Korean, reads \"경주 지진 부상자 23명 · 재산피해 1,118건,\" which translates to \"Gyeongju earthquake, 23 injured, 1,118 property damages.\" This frame continues to illustrate the aftermath of the earthquake, emphasizing the extent of the damage and the affected community."
] |
w6jdwGYcKZU
|
The video depicts the aftermath of an earthquake in Gyeongju, showing widespread destruction and damage to structures. The earthquake resulted in 23 injuries and 1,118 instances of property damage. The footage includes scenes of debris-filled environments, cracked walls, and damaged traditional Korean houses, highlighting the extent of the disaster. The video appears to be part of a news segment, providing visual evidence of the earthquake's impact on the community.
|
The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.
|
[
"People running out of buildings as the earthquake strikes in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Buildings crumbling or collapsing due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"Emergency responders searching for trapped people in rubble after the 2017 Pohang earthquake in Heunghae, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in open spaces after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks forming in roads and sidewalks as the ground shakes during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings crumbling and collapsing during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"People running out of buildings and seeking safety during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders searching for survivors in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cars and debris scattered on the streets after earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in safe zones after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings shaking and crumbling due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"People running for cover and seeking safety outdoors during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders rushing to the scene in ambulances and fire trucks during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Rescue teams searching for trapped people in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks and damage to roads and buildings caused by the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2017%EB%85%84_%ED%8F%AC%ED%95%AD_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: The 2017 Pohang earthquake is a significant event that would have caused destruction and disruption in the affected area, so it makes sense to extract events that are directly related to the earthquake's impact, such as damage to buildings and infrastructure, and the response of emergency services. These events would be concrete and visual, making them suitable for a video caption.\n\nEVENTS:\n1. Buildings shaking and crumbling due to the earthquake\n2. People running for cover and seeking safety outdoors\n3. Emergency responders rushing to the scene in ambulances and fire trucks\n4. Rescue teams searching for trapped people in rubble\n5. Cracks and damage to roads and buildings",
"en_article_excerpt": "The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.",
"en_article_url": "https://en.wikipedia.org/wiki/2017_Pohang_earthquake",
"event_category": "disasters",
"event_name": "pohang_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the immediate consequences and aftermath of the earthquake, as these would be the most visually striking and relevant to a video about the 2017 Pohang earthquake. Events such as damage to buildings, rescue efforts, and people reacting to the earthquake would be likely to be seen in a video associated with the query.\n\nEVENTS:\n1. People running out of buildings as the earthquake strikes\n2. Buildings crumbling or collapsing due to the earthquake\n3. Emergency responders searching for trapped people in rubble\n4. Residents evacuating the area and gathering in open spaces\n5. Cracks forming in roads and sidewalks as the ground shakes",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2017 Pohang earthquake, it makes sense to extract events that are directly related to the aftermath and impact of the earthquake, as these would be concrete and visualizable in a video. The events should reflect the typical consequences of an earthquake, such as damage to infrastructure, rescue efforts, and public response, tailored to the specific context of Pohang, South Korea.\n\nEVENTS:\n1. Buildings crumbling and collapsing\n2. People running out of buildings and seeking safety\n3. Emergency responders searching for survivors in rubble\n4. Cars and debris scattered on the streets\n5. Residents evacuating the area and gathering in safe zones",
"video_URL": "http://youtube.com/watch?v=oK2O8Xtkrvs",
"video_description": "15일 오후 2시 29분쯤 경북 포항시 북구 북쪽 9㎞ 지역에서 규모 5.4의 지진이 발생했다. 오마이TV는 지진 발생 당시 포항시 북구 한살림협동조합 매장을 촬영한 CCTV영상을 제보받아 공개한다. 이 영상에는 갑자기 진열대 물건들이 쏟아져내리는 장면이 들어있다. 한살림협동조합에 근무하는 박혜령(경북 영덕군)씨는 영상을 제보하면서 \"당시 사무실에 있었는데 벽지와 창문이 20초 이상 심하게 흔들렸다\"며 \"아직도 가슴이 진정되지 않는다\"고 말했다. 그는 \"바닥이 계속 흔들려 모두 밖으로 나왔고 안으로 들어갈 엄두를 내지 못했다\"며 \"이후 여진을 알리는 재난문자가 세 차례 왔다\"고 당시 상황을 설명했다. (영상제보 : 박혜령 씨/ 영상편집 : 황지희 기자)",
"video_id": "oK2O8Xtkrvs"
}
| 4
|
[
"In the current frame, a person is crouching near a shelf inside a store. The store has various shelves stocked with products. The text on the screen reads: \"[시민 제보] 포항 지진 순간, 쏟아진 진열대,\" which translates to \"[Citizen Report] Moment of the Pohang earthquake, toppled display.\" This suggests the image captures a moment of disarray likely due to an earthquake.",
"The scene continues to show the aftermath of the earthquake within the store, with several items scattered on the floor. A person stands in the background, appearing to assess the situation. The text \"OhmyTV\" is visible, indicating the source of the footage. The store is in disarray, with shelves and items displaced, highlighting the impact of the earthquake.",
"The video frame shows a continuation of the aftermath within the store, with the text \"OhmyTV\" still visible, reinforcing the source of the footage. The scene remains in disarray, with items and possibly debris scattered across the floor. The person in the background appears to be continuing their assessment of the situation, indicating ongoing efforts to understand the extent of the damage caused by the earthquake.",
"The video frame shows the logo of OhmyTV, indicating it is the source of the footage. The text \"OhmyTV\" is visible, along with a URL and additional Korean text. This frame appears to be the end or a transition screen, providing information about the source and context of the video. The previous frame's description of an aftermath within a store and assessment efforts does not apply here, as this frame is likely unrelated to the store scene and serves a different purpose, such as crediting the footage or providing informational content."
] |
oK2O8Xtkrvs
|
A video captures the moment of and immediate aftermath of the Pohang earthquake inside a store. The footage, sourced from OhmyTV, shows a person crouching near a shelf amidst a disarrayed store with various products scattered on the floor due to the earthquake. As the video progresses, it becomes clear that the store suffered significant damage, with shelves toppled and items displaced. A person in the background is seen assessing the situation, indicating efforts to understand the extent of the damage. The video concludes with a screen displaying the OhmyTV logo, URL, and additional information, serving as a credit or informational segment about the footage.
|
The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.
|
[
"People running out of buildings as the earthquake strikes in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Buildings crumbling or collapsing due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"Emergency responders searching for trapped people in rubble after the 2017 Pohang earthquake in Heunghae, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in open spaces after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks forming in roads and sidewalks as the ground shakes during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings crumbling and collapsing during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"People running out of buildings and seeking safety during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders searching for survivors in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cars and debris scattered on the streets after earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in safe zones after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings shaking and crumbling due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"People running for cover and seeking safety outdoors during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders rushing to the scene in ambulances and fire trucks during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Rescue teams searching for trapped people in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks and damage to roads and buildings caused by the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2017%EB%85%84_%ED%8F%AC%ED%95%AD_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: The 2017 Pohang earthquake is a significant event that would have caused destruction and disruption in the affected area, so it makes sense to extract events that are directly related to the earthquake's impact, such as damage to buildings and infrastructure, and the response of emergency services. These events would be concrete and visual, making them suitable for a video caption.\n\nEVENTS:\n1. Buildings shaking and crumbling due to the earthquake\n2. People running for cover and seeking safety outdoors\n3. Emergency responders rushing to the scene in ambulances and fire trucks\n4. Rescue teams searching for trapped people in rubble\n5. Cracks and damage to roads and buildings",
"en_article_excerpt": "The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.",
"en_article_url": "https://en.wikipedia.org/wiki/2017_Pohang_earthquake",
"event_category": "disasters",
"event_name": "pohang_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the immediate consequences and aftermath of the earthquake, as these would be the most visually striking and relevant to a video about the 2017 Pohang earthquake. Events such as damage to buildings, rescue efforts, and people reacting to the earthquake would be likely to be seen in a video associated with the query.\n\nEVENTS:\n1. People running out of buildings as the earthquake strikes\n2. Buildings crumbling or collapsing due to the earthquake\n3. Emergency responders searching for trapped people in rubble\n4. Residents evacuating the area and gathering in open spaces\n5. Cracks forming in roads and sidewalks as the ground shakes",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2017 Pohang earthquake, it makes sense to extract events that are directly related to the aftermath and impact of the earthquake, as these would be concrete and visualizable in a video. The events should reflect the typical consequences of an earthquake, such as damage to infrastructure, rescue efforts, and public response, tailored to the specific context of Pohang, South Korea.\n\nEVENTS:\n1. Buildings crumbling and collapsing\n2. People running out of buildings and seeking safety\n3. Emergency responders searching for survivors in rubble\n4. Cars and debris scattered on the streets\n5. Residents evacuating the area and gathering in safe zones",
"video_URL": "http://youtube.com/watch?v=6HhGFKbHFbs",
"video_description": "경북 포항에서 규모 5.4의 지진과 여진이 발생, 전국이 지진파에 흔들거리면서 주민들이 불안에 떨고 있다. 15일 기상청에 따르면 이날 오후 2시29분쯤 포항시 북구 흥해 6㎞ 지점에서 지진이 발생했다. http://www.hankookilbo.com/v/38f7f1c1597443e5ac55aca087a5dae2",
"video_id": "6HhGFKbHFbs"
}
| 4
|
[
"The frame shows a street scene with pedestrians walking and a white car parked in the center of the road. Various shops line the street, and a sign with Korean text is visible. The date on the sign indicates November 15.",
"The scene continues on the same street, now with a white arrow painted on the road and a person walking down the middle. The white car is no longer visible. The shops and the sign with Korean text remain unchanged, still indicating November 15. The pedestrian in the center appears to be looking at their phone.",
"The scene remains on the same street, with the person previously seen walking down the middle now walking to the right side of the road. The shops and the sign with Korean text remain unchanged, still indicating November 15. The pedestrian in the center appears to be looking at their phone and is now closer to the crowd of pedestrians on the left.",
"The screen displays a white background with green text reading \"HANKOOKILBO.COM\" and smaller Korean text below it, which indicates a web address and a publication related to Korean news or media."
] |
6HhGFKbHFbs
|
A video depicts a scene on a street in a Korean town, showing pedestrians walking along the road lined with various shops. The date November 15 is visible on a sign. A white car is initially parked in the center of the road but disappears in subsequent frames. A person is seen walking down the middle of the road, looking at their phone, before moving to the right side towards a crowd of pedestrians. The video then cuts to a screen displaying the logo and web address of a Korean news or media publication, HANKOOKILBO.COM, suggesting the footage may be part of a news report or related content.
|
The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.
|
[
"People running out of buildings as the earthquake strikes in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Buildings crumbling or collapsing due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"Emergency responders searching for trapped people in rubble after the 2017 Pohang earthquake in Heunghae, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in open spaces after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks forming in roads and sidewalks as the ground shakes during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings crumbling and collapsing during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"People running out of buildings and seeking safety during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders searching for survivors in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cars and debris scattered on the streets after earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in safe zones after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings shaking and crumbling due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"People running for cover and seeking safety outdoors during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders rushing to the scene in ambulances and fire trucks during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Rescue teams searching for trapped people in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks and damage to roads and buildings caused by the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2017%EB%85%84_%ED%8F%AC%ED%95%AD_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: The 2017 Pohang earthquake is a significant event that would have caused destruction and disruption in the affected area, so it makes sense to extract events that are directly related to the earthquake's impact, such as damage to buildings and infrastructure, and the response of emergency services. These events would be concrete and visual, making them suitable for a video caption.\n\nEVENTS:\n1. Buildings shaking and crumbling due to the earthquake\n2. People running for cover and seeking safety outdoors\n3. Emergency responders rushing to the scene in ambulances and fire trucks\n4. Rescue teams searching for trapped people in rubble\n5. Cracks and damage to roads and buildings",
"en_article_excerpt": "The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.",
"en_article_url": "https://en.wikipedia.org/wiki/2017_Pohang_earthquake",
"event_category": "disasters",
"event_name": "pohang_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the immediate consequences and aftermath of the earthquake, as these would be the most visually striking and relevant to a video about the 2017 Pohang earthquake. Events such as damage to buildings, rescue efforts, and people reacting to the earthquake would be likely to be seen in a video associated with the query.\n\nEVENTS:\n1. People running out of buildings as the earthquake strikes\n2. Buildings crumbling or collapsing due to the earthquake\n3. Emergency responders searching for trapped people in rubble\n4. Residents evacuating the area and gathering in open spaces\n5. Cracks forming in roads and sidewalks as the ground shakes",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2017 Pohang earthquake, it makes sense to extract events that are directly related to the aftermath and impact of the earthquake, as these would be concrete and visualizable in a video. The events should reflect the typical consequences of an earthquake, such as damage to infrastructure, rescue efforts, and public response, tailored to the specific context of Pohang, South Korea.\n\nEVENTS:\n1. Buildings crumbling and collapsing\n2. People running out of buildings and seeking safety\n3. Emergency responders searching for survivors in rubble\n4. Cars and debris scattered on the streets\n5. Residents evacuating the area and gathering in safe zones",
"video_URL": "http://youtube.com/watch?v=1sXMvTYhnn0",
"video_description": "+--------------------------------------------------------------------------------+ ◈ 중앙일보 공식 SNS 채널 ◈ - 페이스북 : https://www.facebook.com/joongang/ - 인스타그램 : https://www.instagram.com/joongangilbo - 트위터 : http://www.twitter.com/joongangilbo - 유튜브 : https://goo.gl/DyGoPC - 네이버 TV : http://tvcast.naver.com/joongangsisa - 네이버포스트 : https://goo.gl/eeYh4k 당신이 원하는 뉴스, 중앙일보 SNS 채널과 함께하세요. +--------------------------------------------------------------------------------+",
"video_id": "1sXMvTYhnn0"
}
| 4
|
[
"The image shows a room with a large grandfather clock on the wall. The clock displays the time as approximately 10:10. Next to it is a smaller wall clock. The background includes bookshelves and various items, including a white helmet. The text on the image is in Korean and discusses an apartment collapse and the unexpected impact of a natural disaster.",
"The image depicts a street scene with a group of people walking along a sidewalk in front of a building. The building has a sign with text in Korean, indicating it is a commercial establishment. There is a parked car visible on the side of the road, and the environment suggests an urban setting. The text at the bottom of the image is in Korean and references a vehicle-related issue, specifically discussing the liability of a driver involved in an accident.",
"The video captures a busy scene at an airport terminal, where several individuals are walking through the departure or arrival area. The text at the bottom in Korean mentions an accident involving a taxi at Daegeu Station, where a driver hit a passenger and is currently being investigated for negligent homicide. The individuals in the image appear to be engaged in various activities, with one person adjusting their mask and another pushing a suitcase, indicating travel or transit activity. The environment is typical of an airport setting, with a mix of travelers and airport staff.",
"The video continues at the airport terminal, with individuals still engaged in travel-related activities. The text at the bottom in Korean continues to detail the taxi accident at Daegeu Station, mentioning the investigation into the driver's actions. The overall setting remains consistent with travelers moving through the area, some adjusting their masks, and others handling their luggage, maintaining the busy atmosphere of the airport."
] |
1sXMvTYhnn0
|
The video depicts a series of unrelated scenes, but the overarching context appears to be news or reportage about various incidents and accidents in an urban setting, likely in Korea given the text in Korean. The video begins in a room with clocks, setting a time context, and transitions to a street scene in front of a commercial building, mentioning a vehicle-related issue. It then shifts to a busy airport terminal, where the focus turns to an accident involving a taxi at Daegeu Station, with a driver under investigation for negligent homicide. Throughout the airport scenes, people are engaged in typical travel activities, oblivious to the news being reported. The key actions involve reporting on accidents, the participants include unnamed individuals in the background of the scenes, and the context is one of urban life interrupted by incidents of negligence or disaster.
|
The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.
|
[
"People running out of buildings as the earthquake strikes in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Buildings crumbling or collapsing due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"Emergency responders searching for trapped people in rubble after the 2017 Pohang earthquake in Heunghae, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in open spaces after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks forming in roads and sidewalks as the ground shakes during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings crumbling and collapsing during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"People running out of buildings and seeking safety during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders searching for survivors in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cars and debris scattered on the streets after earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in safe zones after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings shaking and crumbling due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"People running for cover and seeking safety outdoors during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders rushing to the scene in ambulances and fire trucks during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Rescue teams searching for trapped people in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks and damage to roads and buildings caused by the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2017%EB%85%84_%ED%8F%AC%ED%95%AD_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: The 2017 Pohang earthquake is a significant event that would have caused destruction and disruption in the affected area, so it makes sense to extract events that are directly related to the earthquake's impact, such as damage to buildings and infrastructure, and the response of emergency services. These events would be concrete and visual, making them suitable for a video caption.\n\nEVENTS:\n1. Buildings shaking and crumbling due to the earthquake\n2. People running for cover and seeking safety outdoors\n3. Emergency responders rushing to the scene in ambulances and fire trucks\n4. Rescue teams searching for trapped people in rubble\n5. Cracks and damage to roads and buildings",
"en_article_excerpt": "The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.",
"en_article_url": "https://en.wikipedia.org/wiki/2017_Pohang_earthquake",
"event_category": "disasters",
"event_name": "pohang_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the immediate consequences and aftermath of the earthquake, as these would be the most visually striking and relevant to a video about the 2017 Pohang earthquake. Events such as damage to buildings, rescue efforts, and people reacting to the earthquake would be likely to be seen in a video associated with the query.\n\nEVENTS:\n1. People running out of buildings as the earthquake strikes\n2. Buildings crumbling or collapsing due to the earthquake\n3. Emergency responders searching for trapped people in rubble\n4. Residents evacuating the area and gathering in open spaces\n5. Cracks forming in roads and sidewalks as the ground shakes",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2017 Pohang earthquake, it makes sense to extract events that are directly related to the aftermath and impact of the earthquake, as these would be concrete and visualizable in a video. The events should reflect the typical consequences of an earthquake, such as damage to infrastructure, rescue efforts, and public response, tailored to the specific context of Pohang, South Korea.\n\nEVENTS:\n1. Buildings crumbling and collapsing\n2. People running out of buildings and seeking safety\n3. Emergency responders searching for survivors in rubble\n4. Cars and debris scattered on the streets\n5. Residents evacuating the area and gathering in safe zones",
"video_URL": "http://youtube.com/watch?v=MH6_TeWDitQ",
"video_description": "포항 북구 북쪽 9Km 지역서 5.4 지진 이어 여진…전국서 감지 2.2, 2.6 전진 이어 3.6 여진도 발생 기상청은 15일 오후 2시 29분 쯤 경북 포항시 북구 북쪽 9㎞ 지역에서 규모 5.4의 지진이 발생했다고 밝혔습니다. 진앙은 북위 36.10도, 동경 129.37도입니다. 기상청은 이 지진의 깊이를 9㎞로 파악했습니다. 기상청은 당초 지진 발생지역과 규모를 포항시 북구 북쪽 6㎞ 지점, 규모 5.5로 발표했으나 잠시 후 수정하는 등 혼선이 이어지고 있습니다. 이날 오후 2시49분에는 포항시 북구 북쪽 7km에서 규모 3.6의 여진도 발생했습니다. 기상청은 \"이동속도가 빠른 지진파(P파)만을 이용해 자동 추정한 정보\"라며 \"여진 등 안전에 주의해야 한다\"고 당부했습니다. 이날 지진으로 경북과 경남은 물론 서울 지역에서도 건물 흔들림이 느껴지는 등 전국 곳곳에서 진동이 감지됐습니다. 이보다 앞서 이 인근에서는 두 차례 지진이 있었다고 기상청은 발표했습니다. 오후 2시 22분 32초 포항시 북구 북쪽 7km 지역(북위 36.11도, 동경 129.36도)에서 규모 2.2의 지진이 난 후 2시 22분 44초에는 비슷한 지점(북위 36.08도, 동경 129.31도)에서 규모 2.6의 지진이 발생했습니다. (끝)",
"video_id": "MH6_TeWDitQ"
}
| 4
|
[
"The image shows a map with various regions labeled in Korean. The central focus is a red target-like symbol, indicating a specific location of interest or an event. The surrounding regions include labels such as \"포항시북구\" (Pohang-si Buk-gu), \"포항시남구\" (Pohang-si Nam-gu), \"경주시\" (Gyeongju-si), and \"영천시\" (Yeongcheon-si).",
"The current frame depicts a busy street scene in what appears to be a populated urban area. Numerous people are standing on a pedestrian crosswalk, with several cars parked and moving in the background. The individuals are dressed in a variety of casual and warm clothing, suggesting a cooler climate. The setting is characterized by a mix of residential and commercial buildings, with some buildings featuring signage in Korean. A prominent blue banner with Korean text is visible in the background, suggesting a possible event or gathering. The overall atmosphere is lively and bustling, indicating a potentially significant event or location of interest in the area.",
"The current frame depicts a street scene in an urban area with debris scattered across the road, indicating some form of disturbance or incident. A few people are present, including one person standing on the sidewalk. The surrounding buildings are residential and commercial, with signage in Korean. The presence of debris suggests that the earlier lively atmosphere has changed to one of disorder or aftermath of an event.",
"The scene continues to show an urban street with significant debris scattered across the road, maintaining the sense of disturbance or incident from the previous frame. The same few people, including one standing on the sidewalk, are still present, indicating that they have not moved significantly. The surrounding residential and commercial buildings with Korean signage remain unchanged. The consistent presence of debris suggests that the event causing the disorder is still ongoing or has left a lasting impact on the area."
] |
MH6_TeWDitQ
|
The video depicts an event occurring in an urban area of South Korea, specifically around Pohang and neighboring cities. It begins with a map highlighting a location of interest, then transitions to a bustling street scene with people and cars, suggesting a significant event or gathering. However, the scene quickly changes to one of disorder, with debris scattered across the road, indicating some form of incident or disturbance. The video concludes with the aftermath of this event, showing the same area with debris still present and a few individuals in the vicinity, suggesting that the impact of the incident is ongoing. The overall context implies that the video is capturing the lead-up, occurrence, and immediate aftermath of an event that disrupts normal urban activity in the region.
|
The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.
|
[
"People running out of buildings as the earthquake strikes in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Buildings crumbling or collapsing due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"Emergency responders searching for trapped people in rubble after the 2017 Pohang earthquake in Heunghae, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in open spaces after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks forming in roads and sidewalks as the ground shakes during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings crumbling and collapsing during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"People running out of buildings and seeking safety during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders searching for survivors in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cars and debris scattered on the streets after earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in safe zones after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings shaking and crumbling due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"People running for cover and seeking safety outdoors during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders rushing to the scene in ambulances and fire trucks during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Rescue teams searching for trapped people in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks and damage to roads and buildings caused by the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2017%EB%85%84_%ED%8F%AC%ED%95%AD_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: The 2017 Pohang earthquake is a significant event that would have caused destruction and disruption in the affected area, so it makes sense to extract events that are directly related to the earthquake's impact, such as damage to buildings and infrastructure, and the response of emergency services. These events would be concrete and visual, making them suitable for a video caption.\n\nEVENTS:\n1. Buildings shaking and crumbling due to the earthquake\n2. People running for cover and seeking safety outdoors\n3. Emergency responders rushing to the scene in ambulances and fire trucks\n4. Rescue teams searching for trapped people in rubble\n5. Cracks and damage to roads and buildings",
"en_article_excerpt": "The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.",
"en_article_url": "https://en.wikipedia.org/wiki/2017_Pohang_earthquake",
"event_category": "disasters",
"event_name": "pohang_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the immediate consequences and aftermath of the earthquake, as these would be the most visually striking and relevant to a video about the 2017 Pohang earthquake. Events such as damage to buildings, rescue efforts, and people reacting to the earthquake would be likely to be seen in a video associated with the query.\n\nEVENTS:\n1. People running out of buildings as the earthquake strikes\n2. Buildings crumbling or collapsing due to the earthquake\n3. Emergency responders searching for trapped people in rubble\n4. Residents evacuating the area and gathering in open spaces\n5. Cracks forming in roads and sidewalks as the ground shakes",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2017 Pohang earthquake, it makes sense to extract events that are directly related to the aftermath and impact of the earthquake, as these would be concrete and visualizable in a video. The events should reflect the typical consequences of an earthquake, such as damage to infrastructure, rescue efforts, and public response, tailored to the specific context of Pohang, South Korea.\n\nEVENTS:\n1. Buildings crumbling and collapsing\n2. People running out of buildings and seeking safety\n3. Emergency responders searching for survivors in rubble\n4. Cars and debris scattered on the streets\n5. Residents evacuating the area and gathering in safe zones",
"video_URL": "http://youtube.com/watch?v=_HN3DpC6m6I",
"video_description": "[현장] 포항 지진 피해 상황은…\"여진 등 안전 주의\" 오늘 오후 2시 29분 쯤 경북 포항시 북구에서 규모 5.4의 지진이 발생했습니다. 이번 지진은 작년 9월 12일 경북 경주시에서 발생한 규모 5.8의 지진에 이어 우리나라에서 일어난 지진 중 역대 두번째 규모입니다. 기상청에 따르면 포항지진 발생 지점은 포항시 북구 북쪽 9㎞, 북위 36.10도, 동경 129.37도입니다. 기상청은 이 지진의 깊이를 9㎞로 파악했습니다. 기상청은 당초 지진 발생지역과 규모를 포항시 북구 북쪽 6㎞ 지점, 규모 5.5로 발표했으나 지진 규모를 하향조정하는 등 혼선을 겪었습니다. 규모 5.4 지진에 앞서 오후 2시 22분 32초 포항시 북구 북쪽 7km 지역에서 규모 2.2, 2시 22분 44초 비슷한 지점(북위 36.08도, 동경 129.31도)에서 규모 2.6의 지진 등 전진이 발생했습니다. 아울러 이날 오후 2시 49분 규모 3.6 지진을 시작으로 오후 3시0분 54초께 규모 2.9 지진, 오후 3시 9분 49초 규모 3.6 지진 등 여진도 이어지고 있습니다. 기상청은 \"이동속도가 빠른 지진파(P파)만을 이용해 자동 추정한 정보\"라며 \"여진 등 안전에 주의해야 한다\"고 당부했습니다. 그러면서 \"계속해서 여진이 발생하고 있어 상황을 면밀히 분석 중\"이라며 \"발생 지점이 내륙 쪽으로 들어가 있어 해일의 가능성은 없는 것으로 판단한다\"고 덧붙였습니다. 이날 지진으로 경북과 경남은 물론 서울 지역에서도 건물 흔들림이 느껴지는 등 전국 곳곳에서 진동이 감지됐습니다. 연합뉴스TV : 02-398-4441(기사문의ㆍ제보) 카톡/라인 jebo23 ▣ 연합뉴스TV 유튜브 채널 구독 https://goo.gl/VuCJMi ▣ 대한민국 뉴스의 시작 연합뉴스TV / Yonhap News TV http://www.yonhapnewstv.co.kr/",
"video_id": "_HN3DpC6m6I"
}
| 4
|
[
"The video shows a parking lot under a covered area with several cars parked, including a black car in the center foreground. The scene captures a quiet moment, with no visible people. The timestamp on the video indicates it was recorded on November 15, 2017, at 14:29:32. Text on the video includes \"연합뉴스TV\" and \"현장영상\" in the upper right corner, suggesting that this is a news footage. Another text \"제보영상\" appears at the bottom right, indicating this might be a viewer-submitted clip.",
"The current frame shows a storefront with a large shattered glass window and scattered glass fragments on the ground, indicating a break-in or vandalism. The storefront has a red banner with Korean text. The text \"연합뉴스TV\" and \"현장영상\" are still visible in the upper right corner, confirming it is a news footage. The timestamp suggests the video was recorded on November 15, 2017.",
"The video shows a group of people gathered outdoors in what appears to be a public space with a large metal arch in the background. The crowd consists of men and women, some dressed in suits, suggesting a formal or organized event. The individuals are standing and walking, with some looking towards the camera and others engaged in conversation. The setting includes trees with autumn foliage, indicating the season. In the lower part of the image, Korean text appears, which includes \"기상청\" and \"여진 등 안전에 주의해야...\" suggesting a message from the Korea Meteorological Administration related to safety precautions, possibly regarding aftershocks or other hazards. The text in the upper right corner, \"연합뉴스TV\" and \"현장영상,\" confirms that this is news footage.",
"The video shows a large crowd of people gathered outdoors in front of a large brick building, which appears to be a stadium or sports facility, as indicated by the presence of a scoreboard or large electronic screen. The crowd consists of many individuals, both men and women, some of whom are wearing casual clothing and others in more formal attire, suggesting a mix of public or organized event participants. The individuals are standing, walking, and interacting with each other, with some looking towards the camera and others engaged in conversation. The setting includes trees with autumn foliage, reinforcing the season. The Korean text \"기상청\" and \"여진 등 안전에 주의해야...\" appears in the lower part of the image, indicating a message from the Korea Meteorological Administration regarding safety precautions, possibly due to aftershocks or other hazards. The text \"연합뉴스TV\" and \"현장영상\" in the upper right corner confirms that this is news footage."
] |
_HN3DpC6m6I
|
The video appears to be a news footage compilation from November 15, 2017, covering various scenes related to an event or events in Korea. The footage begins with a quiet parking lot scene, followed by a storefront with a shattered glass window, suggesting a break-in or vandalism. The video then transitions to a public space with a crowd of people, including those dressed in formal attire, near a metal arch and trees with autumn foliage. Finally, it shows a large crowd gathered in front of a stadium or sports facility, with a mix of casual and formal attire, and messages from the Korea Meteorological Administration about safety precautions due to possible aftershocks or hazards. The overall context suggests that the video is reporting on the aftermath of an earthquake or a significant event that has drawn public gatherings and necessitates safety warnings.
|
The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.
|
[
"People running out of buildings as the earthquake strikes in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Buildings crumbling or collapsing due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"Emergency responders searching for trapped people in rubble after the 2017 Pohang earthquake in Heunghae, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in open spaces after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks forming in roads and sidewalks as the ground shakes during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings crumbling and collapsing during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"People running out of buildings and seeking safety during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders searching for survivors in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cars and debris scattered on the streets after earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in safe zones after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings shaking and crumbling due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"People running for cover and seeking safety outdoors during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders rushing to the scene in ambulances and fire trucks during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Rescue teams searching for trapped people in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks and damage to roads and buildings caused by the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2017%EB%85%84_%ED%8F%AC%ED%95%AD_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: The 2017 Pohang earthquake is a significant event that would have caused destruction and disruption in the affected area, so it makes sense to extract events that are directly related to the earthquake's impact, such as damage to buildings and infrastructure, and the response of emergency services. These events would be concrete and visual, making them suitable for a video caption.\n\nEVENTS:\n1. Buildings shaking and crumbling due to the earthquake\n2. People running for cover and seeking safety outdoors\n3. Emergency responders rushing to the scene in ambulances and fire trucks\n4. Rescue teams searching for trapped people in rubble\n5. Cracks and damage to roads and buildings",
"en_article_excerpt": "The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.",
"en_article_url": "https://en.wikipedia.org/wiki/2017_Pohang_earthquake",
"event_category": "disasters",
"event_name": "pohang_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the immediate consequences and aftermath of the earthquake, as these would be the most visually striking and relevant to a video about the 2017 Pohang earthquake. Events such as damage to buildings, rescue efforts, and people reacting to the earthquake would be likely to be seen in a video associated with the query.\n\nEVENTS:\n1. People running out of buildings as the earthquake strikes\n2. Buildings crumbling or collapsing due to the earthquake\n3. Emergency responders searching for trapped people in rubble\n4. Residents evacuating the area and gathering in open spaces\n5. Cracks forming in roads and sidewalks as the ground shakes",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2017 Pohang earthquake, it makes sense to extract events that are directly related to the aftermath and impact of the earthquake, as these would be concrete and visualizable in a video. The events should reflect the typical consequences of an earthquake, such as damage to infrastructure, rescue efforts, and public response, tailored to the specific context of Pohang, South Korea.\n\nEVENTS:\n1. Buildings crumbling and collapsing\n2. People running out of buildings and seeking safety\n3. Emergency responders searching for survivors in rubble\n4. Cars and debris scattered on the streets\n5. Residents evacuating the area and gathering in safe zones",
"video_URL": "http://youtube.com/watch?v=bADRfbUZewY",
"video_description": "산림청은 오는 4일 포항산림조합 목재유통센터에서 국유림에서 생산된 임산 연료를 지진피해를 입은 포항 주민들에게 무상으로 공급하는 행사를 갖기로 했습니다. ◀ＥＮＤ▶ 지진피해 주민과 홀몸노인, 기초생활보장 수급자 등을 대상으로 가구당 1톤씩 모두 21가구를 지원합니다. 이번에 공급하는 장작은 봉화군 장군봉 선도산림경영단지 내 금강소나무 육성사업지에서 생산된 참나무와 벚나무, 박달나무 등입니다.",
"video_id": "bADRfbUZewY"
}
| 4
|
[
"A group of individuals in red uniforms and helmets are handling logs outdoors. They are working near a pile of logs, with some logs being placed into a vehicle. The setting appears to be a community or rural area, and there is a building in the background. The text in Korean indicates assistance provided by Sanlim City and the Police Fire Department.",
"The group of individuals in red uniforms and helmets is actively moving logs. They are seen carrying the logs, with some placing them into a vehicle. The setting remains consistent, indicating a community or rural area, with a building and a gate in the background. The Korean text continues to suggest that the assistance from Sanlim City and the Police Fire Department is ongoing.",
"A person in a red jacket is standing in an outdoor area, amidst a large pile of logs. The setting appears to be a continuation of the community or rural environment from the previous frames, with a building and clear skies in the background. The Korean text indicates that Sanlim City and the Police Fire Department are providing support to the local area. The individual's presence near the logs suggests they are involved in organizing or managing the woodpile.",
"The scene depicts a group of individuals in red jackets and helmets actively engaged in managing and handling logs. The setting remains consistent with the previous frames, indicating a community or rural environment with clear skies. The group appears to be working together, some handling the logs while others observe or assist. The Korean text on the image indicates ongoing support from Sanlim City and the Police Fire Department to the local area, suggesting that this activity is part of a coordinated effort to organize or manage the woodpile."
] |
bADRfbUZewY
|
A group of individuals from Sanlim City and the Police Fire Department, wearing red uniforms and helmets, are working together in a community or rural area to manage and organize a large pile of logs. They are seen carrying and placing logs into a vehicle, with some individuals overseeing the operation. The effort is part of a coordinated assistance program provided by Sanlim City and the Police Fire Department to support the local area. The activity takes place outdoors, near a building, with clear skies in the background, indicating a collaborative community service event focused on wood management.
|
The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.
|
[
"People running out of buildings as the earthquake strikes in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Buildings crumbling or collapsing due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"Emergency responders searching for trapped people in rubble after the 2017 Pohang earthquake in Heunghae, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in open spaces after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks forming in roads and sidewalks as the ground shakes during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings crumbling and collapsing during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"People running out of buildings and seeking safety during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders searching for survivors in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cars and debris scattered on the streets after earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in safe zones after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings shaking and crumbling due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"People running for cover and seeking safety outdoors during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders rushing to the scene in ambulances and fire trucks during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Rescue teams searching for trapped people in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks and damage to roads and buildings caused by the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2017%EB%85%84_%ED%8F%AC%ED%95%AD_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: The 2017 Pohang earthquake is a significant event that would have caused destruction and disruption in the affected area, so it makes sense to extract events that are directly related to the earthquake's impact, such as damage to buildings and infrastructure, and the response of emergency services. These events would be concrete and visual, making them suitable for a video caption.\n\nEVENTS:\n1. Buildings shaking and crumbling due to the earthquake\n2. People running for cover and seeking safety outdoors\n3. Emergency responders rushing to the scene in ambulances and fire trucks\n4. Rescue teams searching for trapped people in rubble\n5. Cracks and damage to roads and buildings",
"en_article_excerpt": "The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.",
"en_article_url": "https://en.wikipedia.org/wiki/2017_Pohang_earthquake",
"event_category": "disasters",
"event_name": "pohang_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the immediate consequences and aftermath of the earthquake, as these would be the most visually striking and relevant to a video about the 2017 Pohang earthquake. Events such as damage to buildings, rescue efforts, and people reacting to the earthquake would be likely to be seen in a video associated with the query.\n\nEVENTS:\n1. People running out of buildings as the earthquake strikes\n2. Buildings crumbling or collapsing due to the earthquake\n3. Emergency responders searching for trapped people in rubble\n4. Residents evacuating the area and gathering in open spaces\n5. Cracks forming in roads and sidewalks as the ground shakes",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2017 Pohang earthquake, it makes sense to extract events that are directly related to the aftermath and impact of the earthquake, as these would be concrete and visualizable in a video. The events should reflect the typical consequences of an earthquake, such as damage to infrastructure, rescue efforts, and public response, tailored to the specific context of Pohang, South Korea.\n\nEVENTS:\n1. Buildings crumbling and collapsing\n2. People running out of buildings and seeking safety\n3. Emergency responders searching for survivors in rubble\n4. Cars and debris scattered on the streets\n5. Residents evacuating the area and gathering in safe zones",
"video_URL": "http://youtube.com/watch?v=4eNANKub4NE",
"video_description": "7000미라클_포항 지진 피해 돕기 특집방송 1부 본방 2017년 11월 28일(화) 오전 10시 10분 재방 2017년 11월 28일(화) 밤 11시 10분 삼방 2017년 11월 29일(수) 오후 2시 50분 ----------------------------------------------------------------------------------------- 7000미라클_포항 지진 피해 돕기 특집방송 2부 본방 2017년 11월 29일(화) 오전 10시 10분 재방 2017년 11월 29일(화) 밤 11시 10분 삼방 2017년 11월 30일(수) 오후 2시 50분",
"video_id": "4eNANKub4NE"
}
| 4
|
[
"The image shows a scene of destruction with a pile of debris including broken concrete blocks, wooden planks, and other construction materials. The background features a partially collapsed structure with a corrugated metal roof and various household items scattered around, suggesting a recent event like a natural disaster or structural collapse. The text in the image mentions \"7000 MIRACLES\" and includes a phone number, indicating that it might be part of a broadcast or a public service announcement.",
"The current frame shows a person sitting inside a car, wearing glasses and a pink jacket with a floral pattern. They appear to be looking out the window, possibly observing something outside. The interior of the car is visible, and there is a logo in the corner of the image indicating that it may be part of a broadcast or public service announcement.",
"The current frame depicts a group of people sitting on the floor, appearing to be in a small, enclosed space, possibly inside a shop or an apartment. The individuals are holding hands in a circle, engaging in what seems to be a moment of solidarity or prayer. There are four visible adults and one child in the circle. The setting includes various household items and clothing, suggesting a domestic environment. A logo in the corner of the image indicates that this scene is part of a broadcast or public service announcement.",
"The current frame shows a concrete cross with a star of David beneath it against a cloudy sky. The cross is part of a larger structure, likely a church or religious building. Text in the upper left corner indicates that the video is related to a broadcast or public service announcement. The presence of religious symbols and the clear sky contrast with the indoor, crowded setting of the previous frame, suggesting a shift from a personal to a broader, possibly spiritual context."
] |
4eNANKub4NE
|
The video depicts a narrative that transitions between scenes of destruction, personal observation, communal solidarity, and spiritual reflection. It begins with a scene of devastation, indicating a recent disaster or structural collapse, and includes a public service announcement element. The story then shifts to a person in a car, observing something outside, suggesting a moment of personal reflection or reaction to the initial scene. The video then cuts to a group of people, including a child, holding hands in a circle, likely in prayer or solidarity, within a domestic setting. Finally, it moves to a broader, spiritual context, showing a religious symbol, a concrete cross with a star of David, against a calm, cloudy sky. The video seems to explore themes of disaster, community, faith, and resilience, possibly as part of a broadcast or public service announcement aimed at inspiring hope or solidarity in the face of adversity.
|
The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.
|
[
"People running out of buildings as the earthquake strikes in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Buildings crumbling or collapsing due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"Emergency responders searching for trapped people in rubble after the 2017 Pohang earthquake in Heunghae, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in open spaces after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks forming in roads and sidewalks as the ground shakes during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings crumbling and collapsing during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"People running out of buildings and seeking safety during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders searching for survivors in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cars and debris scattered on the streets after earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in safe zones after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings shaking and crumbling due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"People running for cover and seeking safety outdoors during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders rushing to the scene in ambulances and fire trucks during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Rescue teams searching for trapped people in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks and damage to roads and buildings caused by the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2017%EB%85%84_%ED%8F%AC%ED%95%AD_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: The 2017 Pohang earthquake is a significant event that would have caused destruction and disruption in the affected area, so it makes sense to extract events that are directly related to the earthquake's impact, such as damage to buildings and infrastructure, and the response of emergency services. These events would be concrete and visual, making them suitable for a video caption.\n\nEVENTS:\n1. Buildings shaking and crumbling due to the earthquake\n2. People running for cover and seeking safety outdoors\n3. Emergency responders rushing to the scene in ambulances and fire trucks\n4. Rescue teams searching for trapped people in rubble\n5. Cracks and damage to roads and buildings",
"en_article_excerpt": "The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.",
"en_article_url": "https://en.wikipedia.org/wiki/2017_Pohang_earthquake",
"event_category": "disasters",
"event_name": "pohang_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the immediate consequences and aftermath of the earthquake, as these would be the most visually striking and relevant to a video about the 2017 Pohang earthquake. Events such as damage to buildings, rescue efforts, and people reacting to the earthquake would be likely to be seen in a video associated with the query.\n\nEVENTS:\n1. People running out of buildings as the earthquake strikes\n2. Buildings crumbling or collapsing due to the earthquake\n3. Emergency responders searching for trapped people in rubble\n4. Residents evacuating the area and gathering in open spaces\n5. Cracks forming in roads and sidewalks as the ground shakes",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2017 Pohang earthquake, it makes sense to extract events that are directly related to the aftermath and impact of the earthquake, as these would be concrete and visualizable in a video. The events should reflect the typical consequences of an earthquake, such as damage to infrastructure, rescue efforts, and public response, tailored to the specific context of Pohang, South Korea.\n\nEVENTS:\n1. Buildings crumbling and collapsing\n2. People running out of buildings and seeking safety\n3. Emergency responders searching for survivors in rubble\n4. Cars and debris scattered on the streets\n5. Residents evacuating the area and gathering in safe zones",
"video_URL": "http://youtube.com/watch?v=QIkpgLcEWMA",
"video_description": "픽업영상|시민들이 보내 준 #포항지진_영상 한동대 외벽 무너지고, 주차장 바닥 갈라져 #JTBC는_여러분들의_제보_기다립니다 ---------------------------------------------------------------------------------------- 📌 소셜스토리 - JTBC 바로가기▼ https://www.facebook.com/JTBCstandbyyou/",
"video_id": "QIkpgLcEWMA"
}
| 4
|
[
"The image shows a damaged building with several broken windows and exposed sections of the interior structure. The text \"소셜스토리\" (Social Story) and \"시청자제보\" (Viewer Report) suggest that this footage is part of a news report from JTBC, likely highlighting the aftermath of an incident or structural damage to the building.",
"In this frame, the scene transitions to an indoor setting where two individuals are seated at a table. They appear to be engaged in a conversation. The table is covered with a red tablecloth and has various items on it, including a large bowl, a cup, and some utensils. One person is gesturing with their hand, possibly explaining something or emphasizing a point. The background features a chalkboard with drawings of stick figures and musical notes. The text \"소셜스토리\" (Social Story) and \"시청자제보\" (Viewer Report) are still present, indicating that this segment continues from the previous one, likely a part of a news or documentary series. The text at the bottom, \"뭐고 이거 진짜진짜,\" suggests that one of the individuals is expressing surprise or disbelief about something.",
"The current frame shows a parking lot with several vehicles, including a white car and a black car. The vehicles are parked closely to each other, suggesting a confined space. The text \"소셜스토리\" (Social Story) and \"시청자제보\" (Viewer Report) are visible, continuing the theme from the previous frames. The scene likely transitions to depict a different aspect of the story being reported, possibly focusing on the setting or an event involving the vehicles.",
"The current frame displays a logo with the text \"JTBC\" against a blue and green gradient background. The logo is centered in the image, suggesting a transition or branding segment within the video, possibly indicating the end of a report or a commercial break."
] |
QIkpgLcEWMA
|
A news report from JTBC, titled "소셜스토리" (Social Story) and "시청자제보" (Viewer Report), depicts the aftermath of an incident involving a damaged building with broken windows and exposed interior structure. The report then transitions to an indoor conversation between two individuals seated at a table, discussing something that prompts surprise or disbelief, indicated by the text "뭐고 이거 진짜진짜." The scene shifts to a parking lot with several vehicles, possibly related to the incident or story being reported. The video concludes with the JTBC logo, suggesting the end of the report or a transition to a different segment.
|
The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.
|
[
"People running out of buildings as the earthquake strikes in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Buildings crumbling or collapsing due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"Emergency responders searching for trapped people in rubble after the 2017 Pohang earthquake in Heunghae, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in open spaces after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks forming in roads and sidewalks as the ground shakes during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings crumbling and collapsing during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"People running out of buildings and seeking safety during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders searching for survivors in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cars and debris scattered on the streets after earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in safe zones after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings shaking and crumbling due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"People running for cover and seeking safety outdoors during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders rushing to the scene in ambulances and fire trucks during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Rescue teams searching for trapped people in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks and damage to roads and buildings caused by the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2017%EB%85%84_%ED%8F%AC%ED%95%AD_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: The 2017 Pohang earthquake is a significant event that would have caused destruction and disruption in the affected area, so it makes sense to extract events that are directly related to the earthquake's impact, such as damage to buildings and infrastructure, and the response of emergency services. These events would be concrete and visual, making them suitable for a video caption.\n\nEVENTS:\n1. Buildings shaking and crumbling due to the earthquake\n2. People running for cover and seeking safety outdoors\n3. Emergency responders rushing to the scene in ambulances and fire trucks\n4. Rescue teams searching for trapped people in rubble\n5. Cracks and damage to roads and buildings",
"en_article_excerpt": "The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.",
"en_article_url": "https://en.wikipedia.org/wiki/2017_Pohang_earthquake",
"event_category": "disasters",
"event_name": "pohang_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the immediate consequences and aftermath of the earthquake, as these would be the most visually striking and relevant to a video about the 2017 Pohang earthquake. Events such as damage to buildings, rescue efforts, and people reacting to the earthquake would be likely to be seen in a video associated with the query.\n\nEVENTS:\n1. People running out of buildings as the earthquake strikes\n2. Buildings crumbling or collapsing due to the earthquake\n3. Emergency responders searching for trapped people in rubble\n4. Residents evacuating the area and gathering in open spaces\n5. Cracks forming in roads and sidewalks as the ground shakes",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2017 Pohang earthquake, it makes sense to extract events that are directly related to the aftermath and impact of the earthquake, as these would be concrete and visualizable in a video. The events should reflect the typical consequences of an earthquake, such as damage to infrastructure, rescue efforts, and public response, tailored to the specific context of Pohang, South Korea.\n\nEVENTS:\n1. Buildings crumbling and collapsing\n2. People running out of buildings and seeking safety\n3. Emergency responders searching for survivors in rubble\n4. Cars and debris scattered on the streets\n5. Residents evacuating the area and gathering in safe zones",
"video_URL": "http://youtube.com/watch?v=_vMNNh-x79Q",
"video_description": "경북 포항서 규모 5.4 지진 발생…서울ㆍ부산서도 진동 [앵커] 오늘 오후 경북 포항에서 규모 5.4의 지진이 발생했습니다. 부산을 비롯해 서울에서도 흔들림이 감지됐습니다. 보도국 취재기자 연결해 자세한 내용 들어보겠습니다. 홍정원 기자. [기자] 네, 오늘 오후 2시29분 포항 북구에서 북쪽으로 6킬로미터 지역에서 규모 5.4의 지진이 발생했다고 기상청이 밝혔습니다. 진앙은 북위 36.10도, 동경 129.37도입니다. 포항에서는 창이 흔들리고 책장에서 책이 떨어지는 등 진동의 여파가 상당했던 것으로 전해지고 있는데요. 또한 포항 곳곳에선 여진도 이어진 것으로 전해지고 있습니다. 기상청은 \"이동속도가 빠른 지진파, P파만을 이용해 자동 추정한 정보\"라며 \"여진 등 안전에 주의해야 한다\"고 당부했습니다. 한수원 측은 전국의 원자력발전소는 별다른 문제 없이 가동 중이라고 밝혔습니다. 오늘 지진의 지진동은 서울 등 경기 남부는 물론, 경남권과 호남권 등 전국적으로 감지된 것으로 전해지고 있습니다. 지진 발생 직후 부산소방안전본부와 부산경찰청에는 건물이 흔들린다는 시민들의 문의 전화가 빗발친 것으로 전해졌습니다. 일부 도심 건물이 심하게 흔들리면서 관련 직장인과 시민들이 급히 대피하는 소동이 벌어졌습니다. 다시 한 번 말씀드리겠습니다. 오늘 오후 2시29분 포항 북구에서 북쪽으로 6킬로미터 지역에서 규모 5.4의 지진이 발생했다고 기상청이 밝혔습니다. 포항 곳곳에선 여진이 이어지면서 주민 대피가 속출했으며, 지진동은 서울 등 경기권과 부산 등에서도 감지되면서 전국적으로 우려를 키우고 있습니다. 원자력발전소를 관리하는 한수원 측은 현재까지 가동중인 원전에 별다른 문제가 없다고 밝혔습니다. 아직까지 전국적으로 정확한 인명, 시설 피해 상황은 집계되지 않은 상태입니다. 지금까지 보도국에서 전해드렸습니다. 연합뉴스TV : 02-398-4441(기사문의·제보) 카톡/라인 jebo23 (끝)",
"video_id": "_vMNNh-x79Q"
}
| 4
|
[
"The image shows an interior view of a room with bright pink blinds and a clock on the wall. There is a person speaking on the phone in the right part of the image. A news broadcast overlay appears, indicating a discussion about economic policies, specifically mentioning '기성형 규모 5.4, 경주 장진 이후 역대 두번째 위력' with additional context including financial data and a stock market index. The video appears to be from a TV news channel, as indicated by the logos and text.",
"The image is a still from a news broadcast, showing a close-up of a blue water bottle with Korean text on its label, placed on a desk. In the background, there are other indistinct office items and a person appears to be speaking on the phone, as indicated by the video overlay. The overlay contains Korean text, likely providing context to the news segment, with references to economic figures and possibly a discussion on policies. The broadcast logo suggests this is from a TV news channel, and the timestamp at the bottom indicates the segment's duration or time.",
"The current frame shows a continuation of the news broadcast, focusing on an interior setting, possibly a living room or kitchen, with a large black dome-shaped light fixture hanging from the ceiling. The overlay text in Korean suggests a discussion about recent events or policies related to economic figures, as indicated by the context from the previous frame. The timestamp at the bottom confirms the continuity of the news segment.",
"The current frame depicts an exterior scene showing a damaged door with broken glass and debris scattered around the threshold. This suggests an incident that has resulted in damage to the property, possibly due to an impact or forced entry. The overlay text in Korean indicates a discussion about the aftermath of a specific event, likely related to the incident shown, mentioning an injury or lack thereof. The presence of a person on the phone in the inset image and the news ticker at the bottom provides context that this is a news report covering recent developments or updates related to the incident. The timestamp confirms the news segment's continuity."
] |
_vMNNh-x79Q
|
A TV news broadcast discusses economic policies and recent events, showing various settings including an interior room with a person on the phone, a close-up of a water bottle on a desk, and an interior with a distinctive light fixture. The broadcast then cuts to an exterior scene of a damaged door with broken glass and debris, indicating coverage of an incident, possibly related to the discussed policies or events. The news segment provides context through overlays with Korean text, including economic figures, policy discussions, and updates on the incident, with a person on the phone in some scenes, suggesting live reporting or analysis.
|
The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.
|
[
"People running out of buildings as the earthquake strikes in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Buildings crumbling or collapsing due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"Emergency responders searching for trapped people in rubble after the 2017 Pohang earthquake in Heunghae, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in open spaces after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks forming in roads and sidewalks as the ground shakes during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings crumbling and collapsing during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"People running out of buildings and seeking safety during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders searching for survivors in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cars and debris scattered on the streets after earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017",
"Residents evacuating the area and gathering in safe zones after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
[
"Buildings shaking and crumbling due to the earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea on 15 November 2017",
"People running for cover and seeking safety outdoors during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Emergency responders rushing to the scene in ambulances and fire trucks during the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Rescue teams searching for trapped people in rubble after the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea",
"Cracks and damage to roads and buildings caused by the 15 November 2017 earthquake in Heunghae, Pohang, North Gyeongsang Province, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2017%EB%85%84_%ED%8F%AC%ED%95%AD_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: The 2017 Pohang earthquake is a significant event that would have caused destruction and disruption in the affected area, so it makes sense to extract events that are directly related to the earthquake's impact, such as damage to buildings and infrastructure, and the response of emergency services. These events would be concrete and visual, making them suitable for a video caption.\n\nEVENTS:\n1. Buildings shaking and crumbling due to the earthquake\n2. People running for cover and seeking safety outdoors\n3. Emergency responders rushing to the scene in ambulances and fire trucks\n4. Rescue teams searching for trapped people in rubble\n5. Cracks and damage to roads and buildings",
"en_article_excerpt": "The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.",
"en_article_url": "https://en.wikipedia.org/wiki/2017_Pohang_earthquake",
"event_category": "disasters",
"event_name": "pohang_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the immediate consequences and aftermath of the earthquake, as these would be the most visually striking and relevant to a video about the 2017 Pohang earthquake. Events such as damage to buildings, rescue efforts, and people reacting to the earthquake would be likely to be seen in a video associated with the query.\n\nEVENTS:\n1. People running out of buildings as the earthquake strikes\n2. Buildings crumbling or collapsing due to the earthquake\n3. Emergency responders searching for trapped people in rubble\n4. Residents evacuating the area and gathering in open spaces\n5. Cracks forming in roads and sidewalks as the ground shakes",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2017 Pohang earthquake, measuring magnitude 5.4 on the Richter magnitude scale, struck Heunghae, Pohang, North Gyeongsang Province, South Korea on November 15, 2017.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2017 Pohang earthquake, it makes sense to extract events that are directly related to the aftermath and impact of the earthquake, as these would be concrete and visualizable in a video. The events should reflect the typical consequences of an earthquake, such as damage to infrastructure, rescue efforts, and public response, tailored to the specific context of Pohang, South Korea.\n\nEVENTS:\n1. Buildings crumbling and collapsing\n2. People running out of buildings and seeking safety\n3. Emergency responders searching for survivors in rubble\n4. Cars and debris scattered on the streets\n5. Residents evacuating the area and gathering in safe zones",
"video_URL": "http://youtube.com/watch?v=k_iUdIJKUUc",
"video_description": "포항 시민들은 눈앞에서 아파트 외벽이 갈라지는 현장을 목격하며 하루종일 공포에 떨었습니다. 시민 700여명은 지금도 집에 돌아가지 못하고 대피소에 머물고 있습니다. [Ch.19] 사실을 보고 진실을 말합니다. ★홈페이지 : http://news.tvchosun.com/ * 뉴스제보 : 이메일(tvchosun@chosun.com), 카카오톡(tv조선제보), 전화(1661-0190)",
"video_id": "k_iUdIJKUUc"
}
| 4
|
[
"A person in a red jacket is standing in a gymnasium with many people sitting on the floor, likely engaged in a communal or educational activity.",
"The current frame shows a damaged building with visible structural cracks and a partially collapsed balcony at night. There is text in Korean at the top and bottom of the image, indicating a news broadcast about the structural damage and the reason for it. The number '9' is visible in the bottom right corner, likely indicating the channel or segment number.",
"In the current frame, a person is seen walking while carrying a blue bag and holding a pink blanket. The person appears to be moving through a dark area, possibly at night. There is text in Korean at the top and bottom of the image, suggesting a news report. The text at the bottom mentions 730 individuals and indicates that this is related to a production or manufacturing process. The number '9' is visible in the bottom right corner, likely indicating the channel or segment number.",
"In the current frame, the person continues walking while carrying the blue bag and holding the pink blanket. The environment remains dark, suggesting it is still nighttime. The Korean text at the top and bottom of the image continues to be present, reinforcing the news report context. The text at the bottom again mentions 730 individuals and the production or manufacturing process. The number '9' is still visible in the bottom right corner, maintaining the channel or segment identification."
] |
k_iUdIJKUUc
|
The video appears to depict a news report or documentary that jumps between different scenes and settings. It starts with a person in a red jacket in a gymnasium, possibly involved in a community or educational activity. The scene then shifts to a damaged building at night, with Korean text indicating a news broadcast about structural damage. The video then follows a person walking at night, carrying a blue bag and holding a pink blanket, with ongoing Korean text suggesting a connection to a production or manufacturing process involving 730 individuals. The consistent presence of the number '9' in the corner of the latter frames suggests these scenes are part of a specific news segment or channel. Overall, the video seems to be a compilation of unrelated or loosely connected events, potentially aiming to highlight different aspects of a community or societal issue.
|
Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Flooding in coastal cities and low-lying areas due to typhoons in the Philippines and Japan",
"Debris-filled streets and damaged buildings after typhoon in Philippines and Japan",
"People evacuating to shelters and higher ground during typhoons in the Philippines and Japan",
"Cars and ships being tossed about by powerful storm surges during typhoons in the Philippines and Japan"
] |
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Heavy rainfall from typhoons causing flooding in urban areas of the Philippines and Japan",
"Storm surges from typhoons crashing against coastal defenses in the Philippines and Japan",
"Debris scattered across roads and streets in Philippines and Japan after typhoon",
"Emergency responders evacuating people from flooded areas in the Philippines and Japan during typhoon events"
] |
[
"Residents in the Philippines and Japan boarding up windows and securing outdoor furniture in preparation for the upcoming typhoon",
"Strong winds and heavy rainfall from typhoon hitting coastal areas of Japan and the Philippines",
"Floodwaters rising in streets and low-lying areas during typhoons in the Philippines and Japan",
"People evacuating to shelters or higher ground during a typhoon in the Philippines and Japan",
"Debris scattered on roads and damage to buildings after a typhoon passes in the Philippines and Japan"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%82%9C%EB%A7%88%EB%8F%8C_(2022%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and scenes associated with a powerful tropical cyclone impacting a region, such as Japan. These events should reflect common visual elements of typhoon coverage, including storm preparations, the storm itself, and its aftermath.\n\nEVENTS:\n1. Residents boarding up windows and securing outdoor furniture in preparation for the storm\n2. Strong winds and heavy rainfall hitting coastal areas of Japan\n3. Floodwaters rising in streets and low-lying areas\n4. People evacuating to shelters or higher ground\n5. Debris scattered on roads and damage to buildings after the storm passes",
"en_article_excerpt": "Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Nanmadol_(2022)",
"event_category": "disasters",
"event_name": "typhoon_nanmadol",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, flooding, and damage to infrastructure, as these are common visual elements that would be captured in a video of the event. Given the location of the typhoon's impact, Japan, the events should be tailored to the country's geography and urban landscape.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. Debris-filled streets and damaged buildings\n4. People evacuating to shelters and higher ground\n5. Cars and ships being tossed about by powerful storm surges",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and aftermath of a powerful tropical cyclone, such as strong winds, heavy rainfall, and resulting damage. These events are likely to be visually captured in a video associated with the query, showcasing the storm's impact on Japan.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Storm surges crashing against coastal defenses\n4. Debris scattered across roads and streets\n5. Emergency responders evacuating people from flooded areas",
"video_URL": "http://youtube.com/watch?v=Be0D3w0LKSg",
"video_description": "(서울=뉴스1) 문동주 기자 = 지난 18일 한국보다 먼저 일본에 상륙한 제14호 태풍 난마돌. 난마돌로 인해 일본에는 엄청난 강풍과 호우 피해가 이어졌다. 사람도 서 있기 힘든 정도의 위력의 비바람에 건물 외장재들이 날아다니고 대중교통 일부도 마비됐다. 동해에 접한 일본 서부 연안을 따라 이동한 난마돌의 위력을 영상으로 확인해보자. #난마돌 #태풍 #일본",
"video_id": "Be0D3w0LKSg"
}
| 4
|
[
"The image shows a scene in Japan experiencing strong winds and a dust storm. The visibility is low due to the dust, and trees in the foreground are bent by the wind. Korean text at the bottom indicates that a strong wind and dust storm are occurring.",
"The video shows the interior of a bus with significant flooding. Water is visibly flowing into the vehicle, reaching the seats and the floor. The bus's dashboard and windows are visible, with rainwater pouring in through the open door. Korean text at the bottom reads \"연천 폭우에 대중교통 일부도 마비,\" which translates to \"Flood in Yeoncheon, public transportation also partially paralyzed.\" This indicates that heavy rainfall has affected public transportation services, leading to disruptions and flooding inside the bus.",
"The video continues to show the interior of the flooded bus, with water still flowing in and reaching the seats and floor. The focus remains on the flooding situation, with the text \"한국도 그렇다\" (Korea is the same) suggesting that the flood issue is widespread in Korea. This reinforces the earlier message that heavy rainfall is affecting public transportation services across the country.",
"The current frame shows a blue screen with the text \"뉴스 TV\" in white, indicating that the scene is transitioning to a news broadcast or segment on a news channel."
] |
Be0D3w0LKSg
|
A severe weather event is affecting Japan and Korea, characterized by strong winds and heavy rainfall. The storm causes a dust storm in Japan with low visibility and bent trees. In Korea, the heavy rainfall leads to flooding, particularly impacting public transportation. A bus is shown with significant flooding, water flowing in through the open door, and reaching the seats and floor, causing disruptions to services. The flooding issue appears to be widespread in Korea, affecting multiple areas, including Yeoncheon. The video then transitions to a news broadcast, suggesting that the footage is part of a news segment covering the severe weather event.
|
Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Flooding in coastal cities and low-lying areas due to typhoons in the Philippines and Japan",
"Debris-filled streets and damaged buildings after typhoon in Philippines and Japan",
"People evacuating to shelters and higher ground during typhoons in the Philippines and Japan",
"Cars and ships being tossed about by powerful storm surges during typhoons in the Philippines and Japan"
] |
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Heavy rainfall from typhoons causing flooding in urban areas of the Philippines and Japan",
"Storm surges from typhoons crashing against coastal defenses in the Philippines and Japan",
"Debris scattered across roads and streets in Philippines and Japan after typhoon",
"Emergency responders evacuating people from flooded areas in the Philippines and Japan during typhoon events"
] |
[
"Residents in the Philippines and Japan boarding up windows and securing outdoor furniture in preparation for the upcoming typhoon",
"Strong winds and heavy rainfall from typhoon hitting coastal areas of Japan and the Philippines",
"Floodwaters rising in streets and low-lying areas during typhoons in the Philippines and Japan",
"People evacuating to shelters or higher ground during a typhoon in the Philippines and Japan",
"Debris scattered on roads and damage to buildings after a typhoon passes in the Philippines and Japan"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%82%9C%EB%A7%88%EB%8F%8C_(2022%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and scenes associated with a powerful tropical cyclone impacting a region, such as Japan. These events should reflect common visual elements of typhoon coverage, including storm preparations, the storm itself, and its aftermath.\n\nEVENTS:\n1. Residents boarding up windows and securing outdoor furniture in preparation for the storm\n2. Strong winds and heavy rainfall hitting coastal areas of Japan\n3. Floodwaters rising in streets and low-lying areas\n4. People evacuating to shelters or higher ground\n5. Debris scattered on roads and damage to buildings after the storm passes",
"en_article_excerpt": "Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Nanmadol_(2022)",
"event_category": "disasters",
"event_name": "typhoon_nanmadol",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, flooding, and damage to infrastructure, as these are common visual elements that would be captured in a video of the event. Given the location of the typhoon's impact, Japan, the events should be tailored to the country's geography and urban landscape.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. Debris-filled streets and damaged buildings\n4. People evacuating to shelters and higher ground\n5. Cars and ships being tossed about by powerful storm surges",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and aftermath of a powerful tropical cyclone, such as strong winds, heavy rainfall, and resulting damage. These events are likely to be visually captured in a video associated with the query, showcasing the storm's impact on Japan.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Storm surges crashing against coastal defenses\n4. Debris scattered across roads and streets\n5. Emergency responders evacuating people from flooded areas",
"video_URL": "http://youtube.com/watch?v=AXoJU3hRho0",
"video_description": "월요일인 오늘(19일)은 태풍 난마돌 의 영향으로 경상권과 강원영동 지역에 강한 비바람이 불겠습니다. 태풍14호 난마돌은 오늘(19일) 오전 3시 현재 일본 가고시마 북북서쪽 약 140km 해상에서 시속 17km로 북진할 것으로 예측되는데요.",
"video_id": "AXoJU3hRho0"
}
| 4
|
[
"The image shows a large water spout or waterspout over a harbor area. In the foreground, fishing rods are visible, indicating the presence of boats or a dock. The Korean text at the bottom of the image indicates that a severe weather event is occurring, with the wind speed reaching approximately 220 km/h, causing the water spout. The setting suggests a stormy maritime environment.",
"The image depicts a geographical map highlighting various regions and water bodies. Several locations are labeled in Korean, such as \"전라남도,\" \"여수,\" and \"대구,\" indicating different areas on the map. The map shows a coastal region with numerous small islands scattered in the sea. The Korean text in the image is likely providing geographical information or directions related to the locations shown. The setting suggests a focus on navigation or exploration of the area, possibly in the context of the previous frame's severe weather event.",
"The current frame shows a clear sky with scattered clouds and a cityscape in the background. The text in Korean on the image reads \"하늘은 대체로 흐리겠지만 저녁 들어 서쪽 지역부터 차차 맑아질것으로 보입니다.\" This translates to \"The sky will be generally cloudy, but it is expected to gradually clear from the western region in the evening.\" The scene suggests a weather update, providing information about the expected change in weather conditions from overcast to clearer skies as the evening progresses.",
"The current frame features a person standing against a plain background. The individual is wearing a black top and has dark hair. The text in Korean on the right side of the image reads \"시청해주셔서 감사합니다.\" which translates to \"Thank you for watching.\" This suggests the conclusion of a segment or video, likely acknowledging the audience's time and attention. The scene is simple and focused on delivering a message of gratitude."
] |
AXoJU3hRho0
|
A severe weather event, characterized by a waterspout with high wind speeds, occurs in a harbor area, affecting the maritime environment. The event is then contextualized within a broader geographical area, shown on a map highlighting various coastal regions and islands. Following the depiction of the storm, a weather update is provided, indicating an expected change in weather conditions from cloudy to clearer skies in the evening. The video concludes with a person expressing gratitude to the audience for watching, marking the end of the segment or video.
|
Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Flooding in coastal cities and low-lying areas due to typhoons in the Philippines and Japan",
"Debris-filled streets and damaged buildings after typhoon in Philippines and Japan",
"People evacuating to shelters and higher ground during typhoons in the Philippines and Japan",
"Cars and ships being tossed about by powerful storm surges during typhoons in the Philippines and Japan"
] |
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Heavy rainfall from typhoons causing flooding in urban areas of the Philippines and Japan",
"Storm surges from typhoons crashing against coastal defenses in the Philippines and Japan",
"Debris scattered across roads and streets in Philippines and Japan after typhoon",
"Emergency responders evacuating people from flooded areas in the Philippines and Japan during typhoon events"
] |
[
"Residents in the Philippines and Japan boarding up windows and securing outdoor furniture in preparation for the upcoming typhoon",
"Strong winds and heavy rainfall from typhoon hitting coastal areas of Japan and the Philippines",
"Floodwaters rising in streets and low-lying areas during typhoons in the Philippines and Japan",
"People evacuating to shelters or higher ground during a typhoon in the Philippines and Japan",
"Debris scattered on roads and damage to buildings after a typhoon passes in the Philippines and Japan"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%82%9C%EB%A7%88%EB%8F%8C_(2022%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and scenes associated with a powerful tropical cyclone impacting a region, such as Japan. These events should reflect common visual elements of typhoon coverage, including storm preparations, the storm itself, and its aftermath.\n\nEVENTS:\n1. Residents boarding up windows and securing outdoor furniture in preparation for the storm\n2. Strong winds and heavy rainfall hitting coastal areas of Japan\n3. Floodwaters rising in streets and low-lying areas\n4. People evacuating to shelters or higher ground\n5. Debris scattered on roads and damage to buildings after the storm passes",
"en_article_excerpt": "Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Nanmadol_(2022)",
"event_category": "disasters",
"event_name": "typhoon_nanmadol",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, flooding, and damage to infrastructure, as these are common visual elements that would be captured in a video of the event. Given the location of the typhoon's impact, Japan, the events should be tailored to the country's geography and urban landscape.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. Debris-filled streets and damaged buildings\n4. People evacuating to shelters and higher ground\n5. Cars and ships being tossed about by powerful storm surges",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and aftermath of a powerful tropical cyclone, such as strong winds, heavy rainfall, and resulting damage. These events are likely to be visually captured in a video associated with the query, showcasing the storm's impact on Japan.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Storm surges crashing against coastal defenses\n4. Debris scattered across roads and streets\n5. Emergency responders evacuating people from flooded areas",
"video_URL": "http://youtube.com/watch?v=MYkG8SmCWNo",
"video_description": "국제우주정거장에서 포착된 제14호 태풍 '난마돌'의 모습이 공개됐습니다. 소용돌이치는 거대한 구름 덩어리 한가운데로 뚜렷한 태풍의 눈이 보입니다. 미국 항공우주국, 나사 소속 우주 비행사인 밥 하인스가 18일, 트위터에 공개한 사진들인데요. 하인스는 국제우주정거장에서 찍은 난마돌의 사진 3장을 올리며, \"태풍 난마돌 경로에 있는 모든 사람들의 안전을 위해 기도한다\"고 적었습니다. https://imnews.imbc.com/replay/2022/nwtoday/article/6408956_35752.html #태풍, #난마돌, #국제우주정거장",
"video_id": "MYkG8SmCWNo"
}
| 4
|
[
"The video features a news anchor standing in front of a newsroom. She is dressed in a maroon top and white skirt, holding a blue pen in her right hand. The background shows a busy office environment with several people working at their desks. There are multiple computer monitors and various office supplies visible. The top of the screen displays text in Korean, indicating news headlines and details about a broadcast, including the time \"19 (월) 6:49\" (19th, Monday, 6:49) and a contact number \"02-784-4000.\" The bottom of the screen also contains additional Korean text, likely providing more news information.",
"The current frame transitions from the news anchor to an aerial view of a large storm cloud. The storm cloud is dense and appears turbulent, with a swirling pattern indicating strong wind activity. The text on the screen in Korean, \"태풍 '난마돌' 일본 상륙 시간당 15km 속도로 북상,\" translates to \"Typhoon 'Nanmadol' lands in Japan, moving north at a speed of 15 km per hour.\" Additionally, the text \"우주에서 본 태풍 '난마돌'\" suggests the storm is being viewed from space. The news program's logo and time \"19 (월) 6:49\" (19th, Monday, 6:49) are still visible, maintaining continuity with the previous frame. The bottom of the screen continues to display more news information, indicating ongoing coverage of the typhoon.",
"The current frame shows a tweet by astronaut Bob Hines, also known as \"Farmer\" Hines, with his Twitter handle @Astro_FarmerBob. The tweet reads, \"It’s incredible how something that seems so beautiful from space can be so terrible on Earth...Praying for the safety of those in the path of Typhoon Nanmadol.\" Accompanying the tweet are three images depicting the typhoon from space, showing the storm's structure and impact. The text on the screen in Korean, \"국제우주정거장서 포착한 난마돌 모습 공개,\" translates to \"View of Nanmadol captured from the International Space Station.\" The news program's logo, time \"19 (월) 6:50\" (19th, Monday, 6:50), and additional news information are still visible, maintaining continuity with the previous frames.",
"The current frame depicts a formal ceremony taking place in an indoor setting, likely a significant public building. Several uniformed individuals are standing in formation on a red carpet, with some holding ceremonial objects like a flag and a sword. The participants are dressed in red and gold uniforms, suggesting a military or ceremonial guard. In the background, there are people walking and observing the event. The Korean text at the top of the frame indicates the event is related to the 'Nanmadol' typhoon, with a specific focus on the '서해5도 강풍주의보' (West Sea 5 islands, strong wind warning). The bottom text provides a sports update, mentioning a promotion of a football team."
] |
MYkG8SmCWNo
|
A news program is broadcasting live, covering the landing of Typhoon Nanmadol in Japan. The news anchor introduces the topic, followed by an aerial view of the storm cloud from space, providing updates on its movement and speed. The program then shares a tweet from astronaut Bob Hines, who expresses concern for those in the typhoon's path and shares images of the storm from the International Space Station. The broadcast then cuts to a formal ceremony, possibly related to the typhoon's impact, where uniformed individuals are participating in a ritual, while the program continues to provide updates on the storm and other news, including a sports report. The overall context is a live news broadcast covering a significant weather event and its potential consequences.
|
Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Flooding in coastal cities and low-lying areas due to typhoons in the Philippines and Japan",
"Debris-filled streets and damaged buildings after typhoon in Philippines and Japan",
"People evacuating to shelters and higher ground during typhoons in the Philippines and Japan",
"Cars and ships being tossed about by powerful storm surges during typhoons in the Philippines and Japan"
] |
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Heavy rainfall from typhoons causing flooding in urban areas of the Philippines and Japan",
"Storm surges from typhoons crashing against coastal defenses in the Philippines and Japan",
"Debris scattered across roads and streets in Philippines and Japan after typhoon",
"Emergency responders evacuating people from flooded areas in the Philippines and Japan during typhoon events"
] |
[
"Residents in the Philippines and Japan boarding up windows and securing outdoor furniture in preparation for the upcoming typhoon",
"Strong winds and heavy rainfall from typhoon hitting coastal areas of Japan and the Philippines",
"Floodwaters rising in streets and low-lying areas during typhoons in the Philippines and Japan",
"People evacuating to shelters or higher ground during a typhoon in the Philippines and Japan",
"Debris scattered on roads and damage to buildings after a typhoon passes in the Philippines and Japan"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%82%9C%EB%A7%88%EB%8F%8C_(2022%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and scenes associated with a powerful tropical cyclone impacting a region, such as Japan. These events should reflect common visual elements of typhoon coverage, including storm preparations, the storm itself, and its aftermath.\n\nEVENTS:\n1. Residents boarding up windows and securing outdoor furniture in preparation for the storm\n2. Strong winds and heavy rainfall hitting coastal areas of Japan\n3. Floodwaters rising in streets and low-lying areas\n4. People evacuating to shelters or higher ground\n5. Debris scattered on roads and damage to buildings after the storm passes",
"en_article_excerpt": "Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Nanmadol_(2022)",
"event_category": "disasters",
"event_name": "typhoon_nanmadol",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, flooding, and damage to infrastructure, as these are common visual elements that would be captured in a video of the event. Given the location of the typhoon's impact, Japan, the events should be tailored to the country's geography and urban landscape.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. Debris-filled streets and damaged buildings\n4. People evacuating to shelters and higher ground\n5. Cars and ships being tossed about by powerful storm surges",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and aftermath of a powerful tropical cyclone, such as strong winds, heavy rainfall, and resulting damage. These events are likely to be visually captured in a video associated with the query, showcasing the storm's impact on Japan.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Storm surges crashing against coastal defenses\n4. Debris scattered across roads and streets\n5. Emergency responders evacuating people from flooded areas",
"video_URL": "http://youtube.com/watch?v=waVeQcHP3f0",
"video_description": "\"우리나라에 미치는 영향은 그대로\"…강한 비바람 태풍 ’난마돌’, 매우 강한 위력 유지한 채 북상 비보다는 바람 강한 태풍…영남 해안에 집중호우 제주도와 영남 해안, 울릉도·독도에 35㎧ 돌풍 [앵커] 북상 중인 14호 태풍 '난마돌'의 진로가 동쪽으로 조금 더 치우치는 것으로 분석됐습니다. 하지만 태풍의 세력이 워낙 강해 제주도와 해안가에는 여전히 비바람 피해가 우려됩니다. 취재 기자 연결해 자세한 태풍 상황 알아보겠습니다. 김진두 기자! 태풍이 일본 쪽으로 더 치우쳐 이동한다는 건데, 그래도 강한 비바람은 그대로 유지된다는 거죠? [기자] 네, 그렇습니다. 기상청은 오늘 오전 브리핑에서 태풍 진로가 미세하게 조정됐다고 밝혔습니다. 태풍이 내일 새벽쯤 일본 규슈 북부에 상륙하는 건 그대로지만, 진로가 약간 더 동쪽으로 움직인다는 겁니다. 우리나라로부터 거리가 더 멀어지는 것이지만, 비바람 예보는 그대로 유지됐습니다. 태풍이 '매우 강한' 위력을 지닌 채 북상하기 때문인데요, 우선 바람이 무척 강하게 불겠고 영남 해안과 영동에는 집중호우도 예상되고 있습니다. 기상청은 제주도와 영남 해안, 그리고 울릉도·독도에 순간 최대 풍속이 35m에 달하는 돌풍이 불겠다고 밝혔습니다. 그 밖의 해안가와 경남 내륙에도 최고 초속 25m, 내륙 지방에도 초속 15m의 순간적인 강풍이 예상됩니다. 비는 영남 해안과 영동지방, 그리고 울릉도, 독도에 50~100mm가 오겠고, 특히 영남 해안에는 150mm가 넘는 호우가 쏟아질 것으로 보입니다. 기상청은 이번 태풍의 고비가 오늘 밤부터 내일 오전 사이가 되겠고, 제주도와 해안가에는 폭풍해일이나 월파 가능성도 있다며 철저한 대비를 당부했습니다. 지금까지 YTN 김진두입니다. YTN 김진두 (jdkim@ytn.co.kr) ※ '당신의 제보가 뉴스가 됩니다' [카카오톡] YTN 검색해 채널 추가 [전화] 02-398-8585 [메일] social@ytn.co.kr ▶ 기사 원문 : https://www.ytn.co.kr/_ln/0108_202209181259028146 ▶ 제보 하기 : https://mj.ytn.co.kr/mj/mj_write.php ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN 무단 전재 및 재배포금지",
"video_id": "waVeQcHP3f0"
}
| 4
|
[
"A man and a woman are sitting at a news desk, presenting information about a tropical storm, as indicated by the weather map in the background showing a storm system. The text on the screen reads in Korean, mentioning that the storm is moving eastward and is expected to affect the area.",
"The scene shows a coastal area affected by the tropical storm, with turbulent waves crashing against rocky shores. Several people are gathered on a wet walkway near the shore, observing the stormy conditions. The text on the screen, in Korean, indicates the storm's impact on the region, likely discussing the severity and effects of the storm on the local area.",
"The frame depicts a fishing harbor severely impacted by strong winds and waves. Several fishing boats are tied to the dock, with their nets and buoys visible, suggesting preparation against the storm. The scene is shrouded in mist, indicating heavy rain or sea spray. The text in Korean translates to \"Fishing boats are struggling in strong winds and waves,\" emphasizing the storm's intensity. The overall atmosphere is one of struggle and resilience against the harsh weather conditions.",
"The frame shows a street scene during heavy rain, with several individuals using umbrellas to shield themselves from the rain. One person is wearing a red backpack and a beige umbrella, while another is dressed in a blue raincoat and holding a blue umbrella with \"Group\" written on it. They are running towards a building entrance, indicating the urgency to find shelter. The background features a tree and a small structure with a yellow roof, suggesting a public or utility building. The atmosphere conveys a sense of urgency and movement amidst the storm."
] |
waVeQcHP3f0
|
A tropical storm is affecting a coastal area, with a news team reporting on its movement and expected impact. The storm brings turbulent waves, strong winds, and heavy rain, causing disruption to the local community. People are seen gathering to observe the storm's effects, while others take shelter from the heavy rain. The fishing industry is also impacted, with boats securing themselves at the harbor. As the storm intensifies, individuals rush to find shelter, highlighting the urgency and challenges posed by the severe weather conditions.
|
Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Flooding in coastal cities and low-lying areas due to typhoons in the Philippines and Japan",
"Debris-filled streets and damaged buildings after typhoon in Philippines and Japan",
"People evacuating to shelters and higher ground during typhoons in the Philippines and Japan",
"Cars and ships being tossed about by powerful storm surges during typhoons in the Philippines and Japan"
] |
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Heavy rainfall from typhoons causing flooding in urban areas of the Philippines and Japan",
"Storm surges from typhoons crashing against coastal defenses in the Philippines and Japan",
"Debris scattered across roads and streets in Philippines and Japan after typhoon",
"Emergency responders evacuating people from flooded areas in the Philippines and Japan during typhoon events"
] |
[
"Residents in the Philippines and Japan boarding up windows and securing outdoor furniture in preparation for the upcoming typhoon",
"Strong winds and heavy rainfall from typhoon hitting coastal areas of Japan and the Philippines",
"Floodwaters rising in streets and low-lying areas during typhoons in the Philippines and Japan",
"People evacuating to shelters or higher ground during a typhoon in the Philippines and Japan",
"Debris scattered on roads and damage to buildings after a typhoon passes in the Philippines and Japan"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%82%9C%EB%A7%88%EB%8F%8C_(2022%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and scenes associated with a powerful tropical cyclone impacting a region, such as Japan. These events should reflect common visual elements of typhoon coverage, including storm preparations, the storm itself, and its aftermath.\n\nEVENTS:\n1. Residents boarding up windows and securing outdoor furniture in preparation for the storm\n2. Strong winds and heavy rainfall hitting coastal areas of Japan\n3. Floodwaters rising in streets and low-lying areas\n4. People evacuating to shelters or higher ground\n5. Debris scattered on roads and damage to buildings after the storm passes",
"en_article_excerpt": "Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Nanmadol_(2022)",
"event_category": "disasters",
"event_name": "typhoon_nanmadol",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, flooding, and damage to infrastructure, as these are common visual elements that would be captured in a video of the event. Given the location of the typhoon's impact, Japan, the events should be tailored to the country's geography and urban landscape.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. Debris-filled streets and damaged buildings\n4. People evacuating to shelters and higher ground\n5. Cars and ships being tossed about by powerful storm surges",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and aftermath of a powerful tropical cyclone, such as strong winds, heavy rainfall, and resulting damage. These events are likely to be visually captured in a video associated with the query, showcasing the storm's impact on Japan.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Storm surges crashing against coastal defenses\n4. Debris scattered across roads and streets\n5. Emergency responders evacuating people from flooded areas",
"video_URL": "http://youtube.com/watch?v=umbAs-igFhc",
"video_description": "역대급 세력을 지닌 14호 태풍 '난마돌'이 일본에 상륙하면서 우리나라 남해와 동해 상은 물론 영남 해안에도 태풍 경보가 내려졌습니다. 현재 부산과 울산 등 태풍 특보 지역에는 초속 30m 안팎의 돌풍이 불고 있어 강풍 피해가 우려됩니다. 또 포항과 경주에는 산사태 주의보가 발령 중 입니다. 기상청은 14호 태풍 '난마돌'이 어제 저녁 일본 가고시마 남쪽에 상륙한 뒤 태풍의 세력이 매우 강에서 강한 강도로 한 단계 약화했다고 밝혔습니다. 하지만 중심 기압은 여전히 955 헥토파스칼로 강한 세력을 유지하고 있고, 중심 부근에는 초속 40m 돌풍이 몰아치고 있습니다. 특히 강풍 반경이 400km로 넓어 우리나라 남해와 동해상은 물론, 부산과 울산, 포항, 경주 등 영남 해안에도 태풍경보가 발령 중입니다. 기상청은 오늘 영남 해안에는 초속 25에서 35m의 돌풍이 불고, 영남 해안과 영동, 울릉도 독도에는 최고 100mm의 호우가 예상된다고 밝혔습니다. 특히 해상에는 최고 10m의 높은 파도가 일겠고, 해안가에는 월파나 폭풍해일 가능성이 크다며 철저한 대비를 당부했습니다. YTN 정혜윤 (jh0302@ytn.co.kr) ※ '당신의 제보가 뉴스가 됩니다' [카카오톡] YTN 검색해 채널 추가 [전화] 02-398-8585 [메일] social@ytn.co.kr ▶ 기사 원문 : https://www.ytn.co.kr/_ln/0108_202209190411440551 ▶ 제보 하기 : https://mj.ytn.co.kr/mj/mj_write.php ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN 무단 전재 및 재배포금지",
"video_id": "umbAs-igFhc"
}
| 4
|
[
"The video shows a nighttime scene on a street. There is a street sign with a white arrow pointing upwards on the left side. A red sign with a blue circle and a white horizontal line, indicating a no entry or no parking zone, is visible on the right. The scene appears to be related to traffic, and there is Korean text at the bottom of the screen, likely providing information about the traffic situation. The text reads: \"현재 부산·울산 등 초속 30m 이상 풍...피해 우려.\"",
"The video shows a nighttime scene on a street. There is a building with a red door on the right side. A red sign with a blue circle and a white horizontal line, indicating a no entry or no parking zone, is visible on the right. The scene appears to be related to traffic, and there is Korean text at the bottom of the screen, likely providing information about the traffic situation. The text reads: \"태풍 세력 다소 약화했지만 중심풍속 초속 40m 강풍\".",
"The video continues to show a nighttime scene, now featuring several firefighters in reflective uniforms and helmets using a flashlight to inspect or work on a section of a building, which appears to be a dark gray or blue color. The firefighters are focused on a specific task, possibly related to the earlier mention of the building with a red door. Korean text at the bottom of the screen reads: \"기상청 영남해안 초속 35m 돌풍, 최고 100mm 호우\", indicating a warning about a storm with strong winds and heavy rain along the southern coast of Korea.",
"The video continues to depict a nighttime scene focused on the emergency response efforts. The firefighters are now working on a section of a building with a white exterior, continuing their task of inspecting or making repairs. The Korean text \"기상청 영남해안 초속 35m 돌풍, 최고 100mm 호우\" remains visible, reinforcing the severe weather conditions. The scene maintains a sense of urgency as the firefighters work diligently under the bright light of their flashlights."
] |
umbAs-igFhc
|
The video depicts a nighttime emergency response scene on a street in South Korea, where firefighters are working to address damage or issues related to a severe storm. The storm, characterized by strong winds of up to 40m/s and heavy rain of up to 100mm, has prompted warnings and emergency measures. The firefighters, dressed in reflective uniforms and helmets, are focused on inspecting and repairing sections of buildings, including one with a red door and another with a white exterior. The scene is set against the backdrop of a traffic-related setting, with signs indicating no entry or parking zones, and Korean text at the bottom of the screen provides updates on the storm's progress and warnings from the weather agency. The overall context suggests that the video is documenting the emergency response efforts to mitigate the impact of the severe weather conditions in the region.
|
Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Flooding in coastal cities and low-lying areas due to typhoons in the Philippines and Japan",
"Debris-filled streets and damaged buildings after typhoon in Philippines and Japan",
"People evacuating to shelters and higher ground during typhoons in the Philippines and Japan",
"Cars and ships being tossed about by powerful storm surges during typhoons in the Philippines and Japan"
] |
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Heavy rainfall from typhoons causing flooding in urban areas of the Philippines and Japan",
"Storm surges from typhoons crashing against coastal defenses in the Philippines and Japan",
"Debris scattered across roads and streets in Philippines and Japan after typhoon",
"Emergency responders evacuating people from flooded areas in the Philippines and Japan during typhoon events"
] |
[
"Residents in the Philippines and Japan boarding up windows and securing outdoor furniture in preparation for the upcoming typhoon",
"Strong winds and heavy rainfall from typhoon hitting coastal areas of Japan and the Philippines",
"Floodwaters rising in streets and low-lying areas during typhoons in the Philippines and Japan",
"People evacuating to shelters or higher ground during a typhoon in the Philippines and Japan",
"Debris scattered on roads and damage to buildings after a typhoon passes in the Philippines and Japan"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%82%9C%EB%A7%88%EB%8F%8C_(2022%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and scenes associated with a powerful tropical cyclone impacting a region, such as Japan. These events should reflect common visual elements of typhoon coverage, including storm preparations, the storm itself, and its aftermath.\n\nEVENTS:\n1. Residents boarding up windows and securing outdoor furniture in preparation for the storm\n2. Strong winds and heavy rainfall hitting coastal areas of Japan\n3. Floodwaters rising in streets and low-lying areas\n4. People evacuating to shelters or higher ground\n5. Debris scattered on roads and damage to buildings after the storm passes",
"en_article_excerpt": "Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Nanmadol_(2022)",
"event_category": "disasters",
"event_name": "typhoon_nanmadol",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, flooding, and damage to infrastructure, as these are common visual elements that would be captured in a video of the event. Given the location of the typhoon's impact, Japan, the events should be tailored to the country's geography and urban landscape.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. Debris-filled streets and damaged buildings\n4. People evacuating to shelters and higher ground\n5. Cars and ships being tossed about by powerful storm surges",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and aftermath of a powerful tropical cyclone, such as strong winds, heavy rainfall, and resulting damage. These events are likely to be visually captured in a video associated with the query, showcasing the storm's impact on Japan.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Storm surges crashing against coastal defenses\n4. Debris scattered across roads and streets\n5. Emergency responders evacuating people from flooded areas",
"video_URL": "http://youtube.com/watch?v=fxLzed6AI2Q",
"video_description": "제14호 태풍 난마돌이 일본 규슈 남부에 상륙했습니다. 일본 기상청은 어제 저녁 7시쯤 태풍이 규슈 남부 가고시마 시 부근에 상륙했다고 발표했습니다. 기상청은 이번 태풍이 규슈 서부 내륙을 따라 북상해 오늘 오전 규슈 북부에서 북동쪽으로 방향을 꺾어 이동할 것으로 예보했습니다. 태풍은 일본 열도를 관통해 오는 21일 오후 홋카이도 동쪽 해역으로 빠져나갈 전망입니다. 일본 기상청은 이번 태풍이 \"지금까지 경험한 적 없는 위험한 태풍\"으로 수십 년에 한 번 있을 만한 대규모 재해가 발생할 가능성이 있다며 각별한 대비를 당부했습니다. 오늘 낮까지 하루 동안 규슈 남부에 최고 500mm, 규슈 북부와 시코쿠에 최고 400mm의 많은 비가 예보돼 있습니다. 한편 가장 먼저 태풍 영향권에 들어선 일본 규슈 남부 지역을 중심으로 10여 명이 강풍에 넘어져 다치는 등 피해가 이어지고 있습니다. 규슈 지역 각 지자체는 주민들을 대상으로 안전한 곳으로 신속히 이동할 것을 권하는 '피난 지시'를 내리는 등 만일의 사태에 대비하고 있습니다. 일본 정부는 재해대책본부를 설치하고 태풍의 움직임과 주민 대피 상황, 그리고 피해 집계 등에 나섰습니다. 기시다 총리는 관계 각료회의를 소집해 국민에게 \"최신 기상 정보와 피난 정보에 주의해 신속히 생명을 지키는 행동을 해달라\"고 촉구했습니다. YTN 이경아 (kalee@ytn.co.kr) ※ '당신의 제보가 뉴스가 됩니다' [카카오톡] YTN 검색해 채널 추가 [전화] 02-398-8585 [메일] social@ytn.co.kr ▶ 기사 원문 : https://www.ytn.co.kr/_ln/0104_202209190251207843 ▶ 제보 하기 : https://mj.ytn.co.kr/mj/mj_write.php ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN 무단 전재 및 재배포금지",
"video_id": "fxLzed6AI2Q"
}
| 4
|
[
"The image shows a stormy or foggy scene with low visibility, where the outline of a hill or landmass is barely visible through the mist. A red banner at the bottom of the image contains Korean text.",
"The image depicts a scene of destruction with debris scattered on the ground, including broken pieces of what appears to be building materials. Yellow caution tape is strung across the area, indicating that it is off-limits. The Korean text at the bottom of the image suggests that the scene is related to a structural collapse, with references to damage and emergency response.",
"The current frame shows a street in a town with heavy rain and low visibility. The Korean text at the bottom indicates that it is a scene of destruction due to a typhoon, with references to collapsed houses and emergency response. The scene is characterized by wet streets, buildings, and utility poles, suggesting recent or ongoing severe weather conditions.",
"The video displays a residential area experiencing severe flooding due to heavy rainfall, likely from a typhoon. The Korean text indicates that the footage was taken from a Twitter account named 'fudosanou'. The scene shows water inundating the ground, partially submerged houses, and trees leaning under the force of strong winds. The environment appears chaotic with low visibility due to rain, highlighting the destructive impact of the storm on the neighborhood."
] |
fxLzed6AI2Q
|
A typhoon has caused widespread destruction and flooding in a residential area, resulting in collapsed houses, debris-filled streets, and severe damage to infrastructure. The storm brought heavy rain and strong winds, leading to a chaotic environment with low visibility. Emergency response efforts are underway, with caution tape marking off-limits areas and references to damage and rescue operations in the accompanying text. The footage, captured from various vantage points, including a Twitter account, showcases the devastating impact of the typhoon on the community, with homes partially submerged in water and trees leaning under the force of the wind.
|
Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Flooding in coastal cities and low-lying areas due to typhoons in the Philippines and Japan",
"Debris-filled streets and damaged buildings after typhoon in Philippines and Japan",
"People evacuating to shelters and higher ground during typhoons in the Philippines and Japan",
"Cars and ships being tossed about by powerful storm surges during typhoons in the Philippines and Japan"
] |
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Heavy rainfall from typhoons causing flooding in urban areas of the Philippines and Japan",
"Storm surges from typhoons crashing against coastal defenses in the Philippines and Japan",
"Debris scattered across roads and streets in Philippines and Japan after typhoon",
"Emergency responders evacuating people from flooded areas in the Philippines and Japan during typhoon events"
] |
[
"Residents in the Philippines and Japan boarding up windows and securing outdoor furniture in preparation for the upcoming typhoon",
"Strong winds and heavy rainfall from typhoon hitting coastal areas of Japan and the Philippines",
"Floodwaters rising in streets and low-lying areas during typhoons in the Philippines and Japan",
"People evacuating to shelters or higher ground during a typhoon in the Philippines and Japan",
"Debris scattered on roads and damage to buildings after a typhoon passes in the Philippines and Japan"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%82%9C%EB%A7%88%EB%8F%8C_(2022%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and scenes associated with a powerful tropical cyclone impacting a region, such as Japan. These events should reflect common visual elements of typhoon coverage, including storm preparations, the storm itself, and its aftermath.\n\nEVENTS:\n1. Residents boarding up windows and securing outdoor furniture in preparation for the storm\n2. Strong winds and heavy rainfall hitting coastal areas of Japan\n3. Floodwaters rising in streets and low-lying areas\n4. People evacuating to shelters or higher ground\n5. Debris scattered on roads and damage to buildings after the storm passes",
"en_article_excerpt": "Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Nanmadol_(2022)",
"event_category": "disasters",
"event_name": "typhoon_nanmadol",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, flooding, and damage to infrastructure, as these are common visual elements that would be captured in a video of the event. Given the location of the typhoon's impact, Japan, the events should be tailored to the country's geography and urban landscape.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. Debris-filled streets and damaged buildings\n4. People evacuating to shelters and higher ground\n5. Cars and ships being tossed about by powerful storm surges",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and aftermath of a powerful tropical cyclone, such as strong winds, heavy rainfall, and resulting damage. These events are likely to be visually captured in a video associated with the query, showcasing the storm's impact on Japan.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Storm surges crashing against coastal defenses\n4. Debris scattered across roads and streets\n5. Emergency responders evacuating people from flooded areas",
"video_URL": "http://youtube.com/watch?v=tVUSMdbETuE",
"video_description": "오늘도 맑은 하늘 보긴 어렵겠습니다. 중국으로 향하는 태풍 '무이파'의 간접영향으로 오전까지 서쪽 지방에는 약한 비가 내리겠는데요, 비의 양이 적어서 낮 동안 활동하시는 데 큰 불편함은 없겠습니다. 때아닌 늦더위도 계속되겠습니다. 오늘 서울이 한낮에 28도, 광주가 31도까지 오르는 등 어제보다 높아 다소 덥겠습니다. 서울 등 수도권은 낮부터 비가 그치고 점차 갤 것으로 보입니다. 오늘 낮 기온은 예년보다 1∼4도가량 높겠고, 특히, 호남 지방을 중심으로 기온이 크게 오르겠습니다. 서울 28도, 전주 30도, 광주 31도가 예상됩니다. 다가오는 주말에는 서울 낮 기온이 30도까지 오르는 등 늦더위가 기승을 부리겠고요, 제주도와 동쪽 지방을 중심으로 비가 내릴 전망입니다. 한편, 오늘 오전 3시쯤, 일본 오키나와 동쪽 해상에서 제14호 태풍 '난마돌'이 발생했습니다. 이 태풍은 세력이 강해지면서 느리게 북상해 다음 주 월요일, 서귀포 남동쪽 약 280km 부근 해상까지 올라오겠습니다. 이후 진로는 유동적이지만, 일본 규슈 부근이나 대한해협으로 이동할 가능성이 있습니다. 앞으로 발표되는 태풍 정보를 잘 확인해주시기 바랍니다. 현재, 태풍 '무이파'의 간접 영향으로, 제주도와 남해안에는 '강풍주의보'가, 대부분 해상에는 '풍랑주의보'가 발효 중입니다. 내륙은 비가 대부분 그쳤지만 동해안과 제주도에는 비가 더 내릴 가능성 있는 만큼 시설물 관리와 해안가 안전사고에 각별히 유의하시기 바랍니다. 지금까지 YTN 원이다입니다. YTN 원이다 (wonleeda95@ytn.co.kr) ※ '당신의 제보가 뉴스가 됩니다' [카카오톡] YTN 검색해 채널 추가 [전화] 02-398-8585 [메일] social@ytn.co.kr ▶ 기사 원문 : https://www.ytn.co.kr/_ln/0108_202209141027402664 ▶ 제보 하기 : https://mj.ytn.co.kr/mj/mj_write.php ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN 무단 전재 및 재배포금지",
"video_id": "tVUSMdbETuE"
}
| 4
|
[
"A news reporter is standing outdoors, holding a microphone with a YTN News logo. She is wearing a purple dress and is speaking to the camera. The background shows trees and a paved area, suggesting an outdoor setting. The text on the screen indicates that it is related to a news broadcast about a 14-hour strike by workers.",
"The image shows a busy city street with numerous pedestrians crossing the road at a crosswalk. The background features modern buildings, suggesting an urban setting. The weather forecast graphic at the top indicates clear to partly cloudy conditions for the next several days, with temperatures ranging from 16°C to 29°C. The text on the screen reads \"오전까지 서쪽 약한 비\" (Weak rain until morning in the west), suggesting a weather update. There is also a notice about a 14-hour strike by workers.",
"The current frame shows a weather forecast graphic with the text \"제14호 태풍 '난마돌' 발생\" (Typhoon 14, 'Nanmadol,' has occurred) indicating the formation of a typhoon. The map displays the projected path of the typhoon, showing its trajectory through various points with timestamps. The image in the background shows rough sea waves, suggesting stormy conditions. The text \"현제\" (Now) indicates the current status of the typhoon.",
"The current frame shows a female news reporter standing outdoors, holding a microphone with the YTN News logo. She is dressed professionally and appears to be reporting live, likely providing an update on the weather forecast or related news coverage. The background includes trees and a vehicle, suggesting an urban setting."
] |
tVUSMdbETuE
|
A news broadcast is covering two main events: a 14-hour strike by workers and the formation of Typhoon Nanmadol. The video begins with a reporter standing outdoors, discussing the worker strike. It then cuts to a busy city street, showing pedestrians and modern buildings, with a weather update on the screen mentioning a strike and a weak rain forecast. The focus shifts to a weather forecast graphic announcing the occurrence of Typhoon Nanmadol, displaying its projected path and current status. The video ends with a female news reporter providing a live update, likely related to the typhoon or the worker strike, in an urban setting with trees and a vehicle in the background.
|
Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Flooding in coastal cities and low-lying areas due to typhoons in the Philippines and Japan",
"Debris-filled streets and damaged buildings after typhoon in Philippines and Japan",
"People evacuating to shelters and higher ground during typhoons in the Philippines and Japan",
"Cars and ships being tossed about by powerful storm surges during typhoons in the Philippines and Japan"
] |
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Heavy rainfall from typhoons causing flooding in urban areas of the Philippines and Japan",
"Storm surges from typhoons crashing against coastal defenses in the Philippines and Japan",
"Debris scattered across roads and streets in Philippines and Japan after typhoon",
"Emergency responders evacuating people from flooded areas in the Philippines and Japan during typhoon events"
] |
[
"Residents in the Philippines and Japan boarding up windows and securing outdoor furniture in preparation for the upcoming typhoon",
"Strong winds and heavy rainfall from typhoon hitting coastal areas of Japan and the Philippines",
"Floodwaters rising in streets and low-lying areas during typhoons in the Philippines and Japan",
"People evacuating to shelters or higher ground during a typhoon in the Philippines and Japan",
"Debris scattered on roads and damage to buildings after a typhoon passes in the Philippines and Japan"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%82%9C%EB%A7%88%EB%8F%8C_(2022%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and scenes associated with a powerful tropical cyclone impacting a region, such as Japan. These events should reflect common visual elements of typhoon coverage, including storm preparations, the storm itself, and its aftermath.\n\nEVENTS:\n1. Residents boarding up windows and securing outdoor furniture in preparation for the storm\n2. Strong winds and heavy rainfall hitting coastal areas of Japan\n3. Floodwaters rising in streets and low-lying areas\n4. People evacuating to shelters or higher ground\n5. Debris scattered on roads and damage to buildings after the storm passes",
"en_article_excerpt": "Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Nanmadol_(2022)",
"event_category": "disasters",
"event_name": "typhoon_nanmadol",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, flooding, and damage to infrastructure, as these are common visual elements that would be captured in a video of the event. Given the location of the typhoon's impact, Japan, the events should be tailored to the country's geography and urban landscape.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. Debris-filled streets and damaged buildings\n4. People evacuating to shelters and higher ground\n5. Cars and ships being tossed about by powerful storm surges",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and aftermath of a powerful tropical cyclone, such as strong winds, heavy rainfall, and resulting damage. These events are likely to be visually captured in a video associated with the query, showcasing the storm's impact on Japan.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Storm surges crashing against coastal defenses\n4. Debris scattered across roads and streets\n5. Emergency responders evacuating people from flooded areas",
"video_URL": "http://youtube.com/watch?v=d6fZyZvF88k",
"video_description": "어젯밤 일본 규슈에 상륙한 태풍 난마돌의 영향으로 수십 명이 다치는 등 피해가 잇따르고 있습니다. NHK는 이번 태풍으로 현재까지 강풍에 넘어져 다치는 등 40여 명이 부상을 입은 것으로 집계됐다고 전했습니다. 이와 함께 태풍 영향권에 있는 지역 주민 약 960만 명에게는 신속한 안전 확보를 요청하는 최고 수준의 재해 경보와 피난 지시 등이 발령됐습니다. 순간최대풍속 50m/s에 이르는 강풍으로 공사용 대형 크레인이 꺾이고 지붕이 날아가는 등 물적 피해도 속출했습니다. 규슈 미야자키에는 지난 15일 이후 1,000mm에 이르는 비가 내리는 등 집중 호우로 인한 도로와 주택 침수도 잇따랐습니다. 많은 비로 하천 수위가 올라가 현재 30여 곳이 범람 위험 수위를 넘어선 것으로 나타났습니다. 일본 국토교통성은 댐 100여 곳의 수문을 미리 열어 범람을 막기 위해 방류를 실시했는데 이런 대규모 방류는 사상 처음입니다. 이와 함께 규슈부터 시코쿠 지역까지 약 31만 가구가 정전됐습니다. 항공과 선박, 철도 등 교통편도 태풍의 영향에서 벗어날 때까지 운행이 대부분 중단됩니다. 규슈 지역을 중심으로 큰 피해를 남긴 태풍 난마돌은 내일 일본 혼슈를 관통해 홋카이도 남쪽 바다로 빠져나갈 전망입니다. 일본 재해 당국은 많은 비가 내린 뒤 토사 붕괴 등 추가 피해가 발생할 가능성에 대비하며 태풍 움직임을 주시하고 있습니다. YTN 이경아 (kalee@ytn.co.kr) ※ '당신의 제보가 뉴스가 됩니다' [카카오톡] YTN 검색해 채널 추가 [전화] 02-398-8585 [메일] social@ytn.co.kr ▶ 기사 원문 : https://www.ytn.co.kr/_ln/0104_202209191047268893 ▶ 제보 하기 : https://mj.ytn.co.kr/mj/mj_write.php ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN 무단 전재 및 재배포금지",
"video_id": "d6fZyZvF88k"
}
| 4
|
[
"The image shows a close-up of a person's feet wearing sandals, standing on a rocky surface near the water. The water appears to be splashing around, suggesting the person might be walking or standing in shallow water. The surrounding environment looks natural and possibly coastal, with rocks visible. There is no visible text in the image.",
"The image shows a nighttime scene, seemingly from a car or vehicle, capturing a road with a fence and a person standing on the side of the road. The person is wearing a reflective vest, indicating they might be a worker or a guide. The headlights of the vehicle illuminate the road and the person, while the surrounding area is dark. The text at the top of the image indicates it is related to a typhoon and provides information about the expected impact of the typhoon, including wind speed and central pressure. The text at the bottom of the image mentions the impact day of the typhoon and the speed of the wind, along with emergency and safety advice.",
"The current frame shows a daytime scene of a damaged roof, seemingly from the aftermath of a typhoon. The roof appears to be severely bent and twisted, with parts of it hanging or completely missing. There are visible structural elements and debris, indicating the force of the wind. The setting appears to be residential, with other buildings visible in the background. The text at the bottom of the image provides information about the typhoon, including its name, and mentions the impact day and the speed of the wind, along with advice on emergency and safety measures.",
"The current frame captures a nighttime scene of a river with high water levels, indicating flooding likely caused by the same typhoon mentioned in the previous frame. The river is swollen, with water levels reaching the edge of a nearby walkway. The walkway, which is illuminated by streetlights, shows signs of water overflow and damage. The surrounding area includes buildings and palm trees, which are swaying in the strong winds. The text overlay on the image provides information about the typhoon's name, the speed of the wind, and emergency contact details, reinforcing the severity of the situation and the need for preparedness."
] |
d6fZyZvF88k
|
A typhoon is impacting a coastal area, causing destruction and flooding. The video begins with a person standing in shallow water near the coast, then cuts to a nighttime scene of a road where a worker or guide is standing, with a vehicle's headlights illuminating the dark surroundings and a text overlay providing information about the typhoon's expected impact. The scene then shifts to a damaged roof in a residential area, showing the aftermath of the typhoon's strong winds. Finally, the video shows a flooded river at night, with water levels reaching the edge of a nearby walkway and surrounding buildings and palm trees swaying in the wind, highlighting the severity of the typhoon's effects. Throughout the video, text overlays provide information about the typhoon, including its name, wind speed, and emergency contact details, emphasizing the need for preparedness and safety measures.
|
Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Flooding in coastal cities and low-lying areas due to typhoons in the Philippines and Japan",
"Debris-filled streets and damaged buildings after typhoon in Philippines and Japan",
"People evacuating to shelters and higher ground during typhoons in the Philippines and Japan",
"Cars and ships being tossed about by powerful storm surges during typhoons in the Philippines and Japan"
] |
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Heavy rainfall from typhoons causing flooding in urban areas of the Philippines and Japan",
"Storm surges from typhoons crashing against coastal defenses in the Philippines and Japan",
"Debris scattered across roads and streets in Philippines and Japan after typhoon",
"Emergency responders evacuating people from flooded areas in the Philippines and Japan during typhoon events"
] |
[
"Residents in the Philippines and Japan boarding up windows and securing outdoor furniture in preparation for the upcoming typhoon",
"Strong winds and heavy rainfall from typhoon hitting coastal areas of Japan and the Philippines",
"Floodwaters rising in streets and low-lying areas during typhoons in the Philippines and Japan",
"People evacuating to shelters or higher ground during a typhoon in the Philippines and Japan",
"Debris scattered on roads and damage to buildings after a typhoon passes in the Philippines and Japan"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%82%9C%EB%A7%88%EB%8F%8C_(2022%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and scenes associated with a powerful tropical cyclone impacting a region, such as Japan. These events should reflect common visual elements of typhoon coverage, including storm preparations, the storm itself, and its aftermath.\n\nEVENTS:\n1. Residents boarding up windows and securing outdoor furniture in preparation for the storm\n2. Strong winds and heavy rainfall hitting coastal areas of Japan\n3. Floodwaters rising in streets and low-lying areas\n4. People evacuating to shelters or higher ground\n5. Debris scattered on roads and damage to buildings after the storm passes",
"en_article_excerpt": "Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Nanmadol_(2022)",
"event_category": "disasters",
"event_name": "typhoon_nanmadol",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, flooding, and damage to infrastructure, as these are common visual elements that would be captured in a video of the event. Given the location of the typhoon's impact, Japan, the events should be tailored to the country's geography and urban landscape.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. Debris-filled streets and damaged buildings\n4. People evacuating to shelters and higher ground\n5. Cars and ships being tossed about by powerful storm surges",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and aftermath of a powerful tropical cyclone, such as strong winds, heavy rainfall, and resulting damage. These events are likely to be visually captured in a video associated with the query, showcasing the storm's impact on Japan.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Storm surges crashing against coastal defenses\n4. Debris scattered across roads and streets\n5. Emergency responders evacuating people from flooded areas",
"video_URL": "http://youtube.com/watch?v=ItDO3MSUQRQ",
"video_description": "힌남노의 상처가 채 가시기도 전에 이번에는 제 14호 태풍 '난마돌'이 북상하고 있습니다. 아직 경로는 유동적이지만 기상청은 다음주 월요일쯤 제주도를 향해 북진할 가능성이 있다고 보고 상황을 예의주시하고 있습니다. #가을태풍#태풍#태풍경로#힌남노#난마돌#태풍난마돌#제14호태풍#기상청#므르복#기상정보#이슈픽#자막뉴스 ☞MBN NEWS 구독하기 : https://www.youtube.com/channel/UCG9aFJTZ-lMCHAiO1KJsirg?sub_confirmation=1 ☞MPLAY 구독하기 : https://www.youtube.com/channel/UC84IlgY5HIzxKBZOt8XM5UA?sub_confirmation=1 ☞MBN Entertainment 구독하기 : https://www.youtube.com/channel/UCYyLIlOJyqkAFKlVjzX5img?sub_confirmation=1 ☞MBN 공식 홈페이지 : https://www.mbn.co.kr",
"video_id": "ItDO3MSUQRQ"
}
| 4
|
[
"The image shows a harbor with calm water, featuring large ropes in the foreground and a dock with several vehicles and a building in the background. The text in the image is in Korean and reads \"우리나라에 영향을 주진 않을 전망,\" which translates to \"It is expected that it will not affect our country.\"",
"The current frame depicts a scene on a city street with debris scattered across the road, suggesting recent destruction or disturbance. The setting appears to be urban, with buildings and streetlights visible in the background. The road is wet, possibly indicating recent rain or the aftermath of a natural disaster. The Korean text overlaid on the image translates to \"It is considered a significant force to be able to launch a 'giant' aircraft.\" This suggests that the video is discussing the capabilities of launching large aircraft and the context of the debris may be related to the discussion on the force required for such launches.",
"The video shows a night scene of a city street with snow falling, creating a dimly lit and somewhat foggy atmosphere. A traffic light is visible with a red signal, and there are buildings and streetlights in the background, indicating an urban setting. The wet road suggests recent snowfall or rain. A weather map is displayed on the right side of the image, showing weather patterns over a map, likely discussing the weather conditions affecting the region. The Korean text translates to \"Wu Jae-hoon, Chief Correspondent, KBS, A low-pressure system is forecast to move towards the west coast of Japan on the 19th, affecting the northern regions,\" indicating that the video is discussing weather forecasts and the impact of a low-pressure system on the region.",
"The current frame shows a dramatic scene of a flooded urban area with high water levels reaching the base of a bridge and a metal fence in the foreground. The weather appears to be extremely foggy, reducing visibility. The water is turbulent, suggesting strong currents or high waves. The presence of the bridge and the fence indicates that the flooding has affected infrastructure and could pose a risk to public safety. The Korean text \"MBN\" is visible, indicating the broadcasting network. This scene likely continues the discussion from the previous frame on severe weather conditions and their impact on the region."
] |
ItDO3MSUQRQ
|
The video depicts a series of scenes related to weather conditions and their impact on urban areas. It begins with a calm harbor scene, followed by a city street with debris, suggesting recent destruction or disturbance, possibly due to a natural disaster. The video then transitions to a night scene with snowfall, discussing weather forecasts and the impact of a low-pressure system on the region. Finally, it shows a dramatic scene of a flooded urban area with high water levels, strong currents, and reduced visibility, indicating severe weather conditions affecting infrastructure and public safety. The overall context suggests a news report or documentary discussing the effects of weather patterns on urban regions, potentially in Korea or Japan, given the Korean text and reference to the west coast of Japan.
|
Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Flooding in coastal cities and low-lying areas due to typhoons in the Philippines and Japan",
"Debris-filled streets and damaged buildings after typhoon in Philippines and Japan",
"People evacuating to shelters and higher ground during typhoons in the Philippines and Japan",
"Cars and ships being tossed about by powerful storm surges during typhoons in the Philippines and Japan"
] |
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines and Japan",
"Heavy rainfall from typhoons causing flooding in urban areas of the Philippines and Japan",
"Storm surges from typhoons crashing against coastal defenses in the Philippines and Japan",
"Debris scattered across roads and streets in Philippines and Japan after typhoon",
"Emergency responders evacuating people from flooded areas in the Philippines and Japan during typhoon events"
] |
[
"Residents in the Philippines and Japan boarding up windows and securing outdoor furniture in preparation for the upcoming typhoon",
"Strong winds and heavy rainfall from typhoon hitting coastal areas of Japan and the Philippines",
"Floodwaters rising in streets and low-lying areas during typhoons in the Philippines and Japan",
"People evacuating to shelters or higher ground during a typhoon in the Philippines and Japan",
"Debris scattered on roads and damage to buildings after a typhoon passes in the Philippines and Japan"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%82%9C%EB%A7%88%EB%8F%8C_(2022%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and scenes associated with a powerful tropical cyclone impacting a region, such as Japan. These events should reflect common visual elements of typhoon coverage, including storm preparations, the storm itself, and its aftermath.\n\nEVENTS:\n1. Residents boarding up windows and securing outdoor furniture in preparation for the storm\n2. Strong winds and heavy rainfall hitting coastal areas of Japan\n3. Floodwaters rising in streets and low-lying areas\n4. People evacuating to shelters or higher ground\n5. Debris scattered on roads and damage to buildings after the storm passes",
"en_article_excerpt": "Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Nanmadol_(2022)",
"event_category": "disasters",
"event_name": "typhoon_nanmadol",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, flooding, and damage to infrastructure, as these are common visual elements that would be captured in a video of the event. Given the location of the typhoon's impact, Japan, the events should be tailored to the country's geography and urban landscape.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. Debris-filled streets and damaged buildings\n4. People evacuating to shelters and higher ground\n5. Cars and ships being tossed about by powerful storm surges",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Nanmadol, known in the Philippines as Super Typhoon Josie, was a powerful tropical cyclone that impacted Japan.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about Typhoon Nanmadol, also known as Super Typhoon Josie, it makes sense to extract events related to the typical effects and aftermath of a powerful tropical cyclone, such as strong winds, heavy rainfall, and resulting damage. These events are likely to be visually captured in a video associated with the query, showcasing the storm's impact on Japan.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Storm surges crashing against coastal defenses\n4. Debris scattered across roads and streets\n5. Emergency responders evacuating people from flooded areas",
"video_URL": "http://youtube.com/watch?v=ZuMemCoZIUY",
"video_description": "북상 중인 제14호 태풍 '난마돌'이 일본 규슈에 상륙할 것으로 보입니다. 기상청은 '난마돌'이 서귀포 남동쪽 먼바다까지 북상한 뒤 급격히 방향을 틀어 다음 주 월요일쯤 일본 규슈에 상륙할 것으로 내다봤습니다. 이후 일본 열도를 따라 계속 북동진할 것으로 보입니다. 따라서 우리나라에 큰 피해를 줄 가능성은 낮아졌지만, 제주도와 남해안, 영남 동해안은 태풍의 직간접적인 영향권에 들어 비바람이 강하게 불 것으로 보입니다. 기상청은 태풍의 진로가 아직은 유동적인 만큼 앞으로 발표되는 태풍 진로에 주의를 기울여 달라고 당부했습니다. YTN 김진두 (jdkim@ytn.co.kr) #태풍난마돌 #난마돌 #규슈 [YTN 사이언스 기사원문] https://science.ytn.co.kr/program/program_view.php?s_mcd=0082&s_hcd=&key=202209151243393568 [프로그램 제작 문의] legbiz@ytn.co.kr",
"video_id": "ZuMemCoZIUY"
}
| 4
|
[
"The image shows a rocky coastline with waves crashing against the rocks. In the background, there are power lines and a structure that appears to be a small building or shelter. The sky is overcast, and the overall atmosphere is misty or foggy. The text at the bottom of the image is in Korean and reads \"기상청 '제14호 태풍 '난마돌' 일본으로 갈듯'\" which translates to \"Meteorological Agency 'Typhoon No. 14 'Namamadole' will likely head towards Japan'\".",
"The image shows a road scene with cars parked and moving along the road. There are palm trees lining the road and a small building with a blue sign in the background. The sky appears overcast, and the overall atmosphere seems calm. The text at the bottom of the image is in Korean and reads \"기상청 '제14호 태풍 '난마돌' 일본으로 갈듯'\" which translates to \"Meteorological Agency 'Typhoon No. 14 'Namamadole' will likely head towards Japan'\".",
"The current frame depicts an outdoor scene with a beach in the background, where a large group of seals is gathered on the sand. In the foreground, there is a structure made of metal beams, possibly part of a viewing platform or barrier. A person is walking towards the left side of the frame, and the sky appears to be overcast, suggesting a cloudy day. The text in the image is in Korean and reads \"제주 남해안 태풍 직격 영향권...강한 비바람 예상\" which translates to \"Jeju South Coast Typhoon Direct Impact Area...Strong Winds and Heavy Rain Expected.\"",
"The current frame shows an outdoor scene near a coastal area. In the background, the ocean is visible with waves crashing onto the shore, and the sky appears overcast, consistent with the previous description of a cloudy day. In the foreground, there is a large statue of a figure with a bag, possibly representing a traveler or adventurer. Another smaller statue is seen to the right, and a structure with a Korean sign is visible in the distance, indicating that this scene is likely in Jeju, South Korea. The text in the image reiterates the earlier message about the direct impact area of a typhoon with strong winds and heavy rain expected."
] |
ZuMemCoZIUY
|
A video depicts the anticipated impact of Typhoon Namamadole on Jeju, South Korea. The scenes show various locations, including a rocky coastline, a road, and a beach with a group of seals, all under overcast skies. The typhoon's approach is expected to bring strong winds and heavy rain to the area, as indicated by text overlays in Korean. The video does not show the storm itself but rather the pre-typhoon conditions and the warning messages from the Meteorological Agency. The setting is primarily outdoors, with structures such as buildings, statues, and possibly a viewing platform visible in the background. The participants are minimal, with only one person seen walking in one of the frames, suggesting the video is more focused on the environmental and weather conditions than on individual actions or interactions. Overall, the video appears to be informative, aiming to alert viewers about the impending typhoon and its potential effects on Jeju.
|
On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.
|
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022",
"Cars and vehicles submerged in floodwater during the Seoul floods in Gangnam District, South Korea on 8 August 2022",
"Emergency responders rescuing individuals from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and property damage from floodwaters in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing out flooded homes and businesses in Seoul's Gangnam District after floods on 8 August 2022"
] |
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022 during floods",
"Cars and buses submerged in floodwater during the 8 August 2022 floods in Seoul's Gangnam District, South Korea",
"Emergency responders rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing debris and damaged belongings from their homes after floods in Gangnam District, Seoul, South Korea on 8 August 2022",
"Floodwaters overflowing from rivers and streams in Gangnam District, Seoul, South Korea during the floods on 8 August 2022"
] |
[
"People wading through floodwaters in the streets of Gangnam District Seoul South Korea during the floods on 8 August 2022",
"Cars and buses submerged in water on flooded roads in Gangnam District, Seoul, South Korea during the floods on August 8, 2022",
"Residents evacuating their homes and seeking higher ground in Seoul, South Korea, Gangnam District due to floods on 8 August 2022",
"Emergency responders, such as firefighters and police, rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and damaged furniture floating in the floodwaters during the 8 August 2022 floods in Gangnam District, Seoul, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2022%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A4%91%EB%B6%80_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District, it makes sense to extract events that are commonly associated with flooding in urban areas, such as people evacuating, floodwaters rising, and damage to infrastructure. These events are likely to be visualized in a video capturing the aftermath of the flood.\n\nEVENTS:\n1. People wading through floodwaters in the streets of Gangnam District\n2. Cars and buses submerged in water on flooded roads\n3. Residents evacuating their homes and seeking higher ground\n4. Emergency responders, such as firefighters and police, rescuing people from flooded areas\n5. Debris and damaged furniture floating in the floodwaters",
"en_article_excerpt": "On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.",
"en_article_url": "https://en.wikipedia.org/wiki/2022_South_Korean_floods",
"event_category": "disasters",
"event_name": "2022_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, particularly the Gangnam District and surrounding areas, it makes sense to extract events that are directly related to flooding and its immediate consequences, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visually captured in a video documenting the floods and their impact on the affected areas.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and vehicles submerged in floodwater\n3. Emergency responders rescuing individuals from flooded areas\n4. Debris and property damage from floodwaters\n5. Residents clearing out flooded homes and businesses",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District and surrounding areas, it makes sense to extract events that are commonly associated with flooding, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visualized in a video capturing the aftermath and response to the flooding in Seoul.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and buses submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Residents clearing debris and damaged belongings from their homes\n5. Floodwaters overflowing from rivers and streams in the affected areas",
"video_URL": "http://twitter.com/i/status/1556640821054787588",
"video_description": "I just found out about the subway floods in Korea, 모두 건강 유의하세요 ! 아무쪼록 인명피해는 없게 해주세요.",
"video_id": "1556640821054787588"
}
| 4
|
[
"A large amount of water is cascading down from the ceiling in a subway station. The area is flooded, with water pooling on the floor, and a yellow tactile strip is visible on the ground. Signs and maps are on the wall, indicating directions and station information.",
"The water continues to pour down from the ceiling in the subway station, contributing to the ongoing flooding on the floor. The yellow tactile strip is still visible, and the wall signs and maps continue to provide station information and directions amidst the waterlogged conditions.",
"In the current frame, the ceiling tiles are visibly caving in, with chunks of the ceiling collapsing and water continuing to pour down into the subway station. The floor remains heavily flooded, and the yellow tactile strip is still present. The wall signs and maps continue to be visible, providing station information and directions amidst the deteriorating conditions.",
"The scene depicts a subway station where the ceiling tiles are continuing to collapse, with additional pieces falling and water pouring down. The floor remains flooded, and the yellow tactile strip is still visible. Wall signs and maps are visible, indicating station information and directions, as the conditions continue to deteriorate."
] |
1556640821054787588
|
A subway station is experiencing a severe flood due to water cascading down from the ceiling, causing the floor to become heavily waterlogged. As the situation worsens, the ceiling tiles begin to cave in and collapse, with chunks falling to the ground. Despite the deteriorating conditions, wall signs and maps remain visible, providing station information and directions. The flooding and ceiling collapse continue to intensify over time, with no indication of intervention or resolution.
|
On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.
|
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022",
"Cars and vehicles submerged in floodwater during the Seoul floods in Gangnam District, South Korea on 8 August 2022",
"Emergency responders rescuing individuals from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and property damage from floodwaters in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing out flooded homes and businesses in Seoul's Gangnam District after floods on 8 August 2022"
] |
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022 during floods",
"Cars and buses submerged in floodwater during the 8 August 2022 floods in Seoul's Gangnam District, South Korea",
"Emergency responders rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing debris and damaged belongings from their homes after floods in Gangnam District, Seoul, South Korea on 8 August 2022",
"Floodwaters overflowing from rivers and streams in Gangnam District, Seoul, South Korea during the floods on 8 August 2022"
] |
[
"People wading through floodwaters in the streets of Gangnam District Seoul South Korea during the floods on 8 August 2022",
"Cars and buses submerged in water on flooded roads in Gangnam District, Seoul, South Korea during the floods on August 8, 2022",
"Residents evacuating their homes and seeking higher ground in Seoul, South Korea, Gangnam District due to floods on 8 August 2022",
"Emergency responders, such as firefighters and police, rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and damaged furniture floating in the floodwaters during the 8 August 2022 floods in Gangnam District, Seoul, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2022%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A4%91%EB%B6%80_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District, it makes sense to extract events that are commonly associated with flooding in urban areas, such as people evacuating, floodwaters rising, and damage to infrastructure. These events are likely to be visualized in a video capturing the aftermath of the flood.\n\nEVENTS:\n1. People wading through floodwaters in the streets of Gangnam District\n2. Cars and buses submerged in water on flooded roads\n3. Residents evacuating their homes and seeking higher ground\n4. Emergency responders, such as firefighters and police, rescuing people from flooded areas\n5. Debris and damaged furniture floating in the floodwaters",
"en_article_excerpt": "On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.",
"en_article_url": "https://en.wikipedia.org/wiki/2022_South_Korean_floods",
"event_category": "disasters",
"event_name": "2022_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, particularly the Gangnam District and surrounding areas, it makes sense to extract events that are directly related to flooding and its immediate consequences, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visually captured in a video documenting the floods and their impact on the affected areas.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and vehicles submerged in floodwater\n3. Emergency responders rescuing individuals from flooded areas\n4. Debris and property damage from floodwaters\n5. Residents clearing out flooded homes and businesses",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District and surrounding areas, it makes sense to extract events that are commonly associated with flooding, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visualized in a video capturing the aftermath and response to the flooding in Seoul.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and buses submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Residents clearing debris and damaged belongings from their homes\n5. Floodwaters overflowing from rivers and streams in the affected areas",
"video_URL": "http://twitter.com/i/status/1556678909436932096",
"video_description": "This is absolutely insane video in South Korea. Extreme flooding that hasn’t been seen in 80 years or more. Very extreme flooding, causing massive problems in Seoul and the entirety of the country. #SouthKorea #flooding #weather",
"video_id": "1556678909436932096"
}
| 4
|
[
"The image shows a flooded urban street with turbulent water covering the road, creating a strong current. A bus is visible in the background, along with a few cars and a traffic light. There is also a person wearing a reflective safety vest standing near the bus stop, likely indicating the presence of emergency or maintenance workers. Text in Korean is displayed at the bottom of the image, likely providing commentary or context about the flooding situation. The location is identified as '사당역' (Sadang Station).",
"The current frame continues to depict the flooded urban street at Sadang Station. The turbulent water still covers the road, and the strong current is visible, with a white car partially submerged and moving through the flood. The bus in the background is still present, along with other vehicles navigating the inundated area. The traffic light is visible, now showing green. The person in the reflective safety vest remains near the bus stop, indicating ongoing emergency or maintenance efforts. The text in Korean at the bottom of the image seems to provide additional commentary or context about the flooding situation, suggesting a narrative of the event's progression and the efforts to manage it.",
"The current frame continues to depict the flooded urban street at Sadang Station. The turbulent water still covers the road, and the strong current is visible, with a white car partially submerged and moving through the flood. The bus in the background is still present, along with other vehicles navigating the inundated area. The traffic light is visible, now showing green. The person in the reflective safety vest remains near the bus stop, indicating ongoing emergency or maintenance efforts. The text in Korean at the bottom of the image seems to provide additional commentary or context about the flooding situation, suggesting a narrative of the event's progression and the efforts to manage it.",
"The current frame continues to depict the flooded urban street at Sadang Station. The turbulent water still covers the road, and the strong current is visible, with a white car partially submerged and moving through the flood. The bus in the background is still present, along with other vehicles navigating the inundated area. The traffic light is visible, now showing green. The person in the reflective safety vest remains near the bus stop, indicating ongoing emergency or maintenance efforts. The text in Korean at the bottom of the image seems to provide additional commentary or context about the flooding situation, suggesting a narrative of the event's progression and the efforts to manage it."
] |
1556678909436932096
|
The video depicts a flooded urban street at Sadang Station, with turbulent water covering the road and creating a strong current. A bus and several cars, including a partially submerged white car, are visible in the scene, along with a traffic light and a person wearing a reflective safety vest near the bus stop, indicating emergency or maintenance efforts. The flooding situation is dynamic, with vehicles navigating the inundated area and the traffic light showing green. The video appears to show a continuous scene of the flooding, with the person in the reflective vest remaining near the bus stop, suggesting ongoing efforts to manage the situation. The text in Korean at the bottom of the frames provides additional commentary or context about the flooding situation, narrating the event's progression and the responses to it.
|
On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.
|
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022",
"Cars and vehicles submerged in floodwater during the Seoul floods in Gangnam District, South Korea on 8 August 2022",
"Emergency responders rescuing individuals from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and property damage from floodwaters in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing out flooded homes and businesses in Seoul's Gangnam District after floods on 8 August 2022"
] |
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022 during floods",
"Cars and buses submerged in floodwater during the 8 August 2022 floods in Seoul's Gangnam District, South Korea",
"Emergency responders rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing debris and damaged belongings from their homes after floods in Gangnam District, Seoul, South Korea on 8 August 2022",
"Floodwaters overflowing from rivers and streams in Gangnam District, Seoul, South Korea during the floods on 8 August 2022"
] |
[
"People wading through floodwaters in the streets of Gangnam District Seoul South Korea during the floods on 8 August 2022",
"Cars and buses submerged in water on flooded roads in Gangnam District, Seoul, South Korea during the floods on August 8, 2022",
"Residents evacuating their homes and seeking higher ground in Seoul, South Korea, Gangnam District due to floods on 8 August 2022",
"Emergency responders, such as firefighters and police, rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and damaged furniture floating in the floodwaters during the 8 August 2022 floods in Gangnam District, Seoul, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2022%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A4%91%EB%B6%80_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District, it makes sense to extract events that are commonly associated with flooding in urban areas, such as people evacuating, floodwaters rising, and damage to infrastructure. These events are likely to be visualized in a video capturing the aftermath of the flood.\n\nEVENTS:\n1. People wading through floodwaters in the streets of Gangnam District\n2. Cars and buses submerged in water on flooded roads\n3. Residents evacuating their homes and seeking higher ground\n4. Emergency responders, such as firefighters and police, rescuing people from flooded areas\n5. Debris and damaged furniture floating in the floodwaters",
"en_article_excerpt": "On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.",
"en_article_url": "https://en.wikipedia.org/wiki/2022_South_Korean_floods",
"event_category": "disasters",
"event_name": "2022_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, particularly the Gangnam District and surrounding areas, it makes sense to extract events that are directly related to flooding and its immediate consequences, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visually captured in a video documenting the floods and their impact on the affected areas.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and vehicles submerged in floodwater\n3. Emergency responders rescuing individuals from flooded areas\n4. Debris and property damage from floodwaters\n5. Residents clearing out flooded homes and businesses",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District and surrounding areas, it makes sense to extract events that are commonly associated with flooding, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visualized in a video capturing the aftermath and response to the flooding in Seoul.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and buses submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Residents clearing debris and damaged belongings from their homes\n5. Floodwaters overflowing from rivers and streams in the affected areas",
"video_URL": "http://twitter.com/i/status/1556973165556903939",
"video_description": "Visuals after yesterday's floods... Rains continue in S. Korea on Tuesday; \"level 3\" warning in operation to prevent damage. Be careful.. stay Safe.. #SouthKorea #flooding #flood #Seoul #climate #rain",
"video_id": "1556973165556903939"
}
| 4
|
[
"The image shows the exterior of a building with a sign in Korean, indicating it may be a business or establishment. The entrance is covered with a transparent plastic sheet, possibly due to weather conditions or renovation. A wooden stool is visible on the left side, and a trash bin is near the door. The surrounding area appears to be tiled.",
"The image depicts a rainy city street at night. Several people are walking, carrying umbrellas to shield themselves from the rain. The ground is wet, reflecting the bright neon lights from the surrounding buildings and billboards. One person is wearing a light blue shirt and a face mask. Another person is dressed in dark clothing and is closer to the camera, holding a black umbrella. The neon lights create a vibrant and lively atmosphere, despite the rain.",
"The video continues on the rainy city street at night, showcasing a wet ground reflecting the neon lights from the buildings. People are still navigating the area, some using umbrellas. A person in light blue, wearing a face mask, is visible among them. Another individual in dark attire holds a black umbrella, remaining close to the camera. Neon lights maintain the vibrant atmosphere, enhancing the scene's lively feel despite the rain.",
"The scene continues on the rainy city street at night, maintaining the vibrant ambiance with neon lights. A white car is parked on the wet road, and a few people are walking under umbrellas. A person in a green coat is walking towards the camera, and another individual in light blue is still visible, wearing a face mask. The atmosphere remains lively and bustling despite the rain, with neon lights from buildings enhancing the scene's brightness."
] |
1556973165556903939
|
The video depicts a rainy night scene in a city, showing people navigating through the wet streets under the vibrant glow of neon lights from surrounding buildings and billboards. The scene is set outside a business or establishment with a Korean sign, though the focus shifts to the street where individuals, some wearing face masks and carrying umbrellas, move about. The atmosphere is lively, with reflections of neon lights on the wet ground adding to the visual dynamism. Various individuals are captured, including those in light blue, dark clothing, and a green coat, as they walk through the rain-soaked streets, passing by a parked white car. Despite the weather, the city street maintains a bustling and vibrant ambiance.
|
On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.
|
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022",
"Cars and vehicles submerged in floodwater during the Seoul floods in Gangnam District, South Korea on 8 August 2022",
"Emergency responders rescuing individuals from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and property damage from floodwaters in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing out flooded homes and businesses in Seoul's Gangnam District after floods on 8 August 2022"
] |
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022 during floods",
"Cars and buses submerged in floodwater during the 8 August 2022 floods in Seoul's Gangnam District, South Korea",
"Emergency responders rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing debris and damaged belongings from their homes after floods in Gangnam District, Seoul, South Korea on 8 August 2022",
"Floodwaters overflowing from rivers and streams in Gangnam District, Seoul, South Korea during the floods on 8 August 2022"
] |
[
"People wading through floodwaters in the streets of Gangnam District Seoul South Korea during the floods on 8 August 2022",
"Cars and buses submerged in water on flooded roads in Gangnam District, Seoul, South Korea during the floods on August 8, 2022",
"Residents evacuating their homes and seeking higher ground in Seoul, South Korea, Gangnam District due to floods on 8 August 2022",
"Emergency responders, such as firefighters and police, rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and damaged furniture floating in the floodwaters during the 8 August 2022 floods in Gangnam District, Seoul, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2022%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A4%91%EB%B6%80_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District, it makes sense to extract events that are commonly associated with flooding in urban areas, such as people evacuating, floodwaters rising, and damage to infrastructure. These events are likely to be visualized in a video capturing the aftermath of the flood.\n\nEVENTS:\n1. People wading through floodwaters in the streets of Gangnam District\n2. Cars and buses submerged in water on flooded roads\n3. Residents evacuating their homes and seeking higher ground\n4. Emergency responders, such as firefighters and police, rescuing people from flooded areas\n5. Debris and damaged furniture floating in the floodwaters",
"en_article_excerpt": "On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.",
"en_article_url": "https://en.wikipedia.org/wiki/2022_South_Korean_floods",
"event_category": "disasters",
"event_name": "2022_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, particularly the Gangnam District and surrounding areas, it makes sense to extract events that are directly related to flooding and its immediate consequences, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visually captured in a video documenting the floods and their impact on the affected areas.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and vehicles submerged in floodwater\n3. Emergency responders rescuing individuals from flooded areas\n4. Debris and property damage from floodwaters\n5. Residents clearing out flooded homes and businesses",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District and surrounding areas, it makes sense to extract events that are commonly associated with flooding, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visualized in a video capturing the aftermath and response to the flooding in Seoul.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and buses submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Residents clearing debris and damaged belongings from their homes\n5. Floodwaters overflowing from rivers and streams in the affected areas",
"video_URL": "http://twitter.com/i/status/1556659852054634496",
"video_description": "Dorimcheon stream in #Seoul's Gwanak District right now. People are now asking where President Yoon Suk-yeol is and what he's doing. Mayor Oh has just returned to office to deal with disaster, but it's not just Seoul. #SouthKorea #flooding #floodsituation",
"video_id": "1556659852054634496"
}
| 4
|
[
"The current frame depicts a nighttime scene of a flooded area. There is a metal railing in the foreground, and water is rushing through a channel beneath it. On the left, there are streetlights illuminating the area, and a hedge is lit by a light source. A building with illuminated signage is visible in the background, and the street appears to be flooded.",
"The scene continues to depict a nighttime view of a flooded area. The metal railing remains in the foreground, with water still flowing rapidly beneath it. The streetlights continue to illuminate the area, and the hedge is still lit by a light source. The building with illuminated signage is visible in the background, but the street now appears to be more heavily flooded, indicating an increase in water levels.",
"The scene shows a nighttime view of a severely flooded area. A metal railing is prominently visible in the foreground, with turbulent water flowing rapidly beneath it. The water level appears to be higher than before, partially submerging the railing and nearby vegetation. Streetlights continue to illuminate the scene, casting a dim glow on the turbulent waters. The visibility is reduced due to the heavy rainfall, with the background obscured by mist and darkness. The overall atmosphere is one of heightened urgency and danger due to the rising floodwaters.",
"The scene remains set in a nighttime flood with a metal railing in the foreground, though the water level has receded slightly. The turbulent water, still flowing rapidly beneath the railing, is less intense than before. Vegetation near the railing is no longer partially submerged, and the streetlights continue to dimly illuminate the area. The visibility has slightly improved, with the background less obscured by mist and darkness. The atmosphere suggests a gradual decrease in the flood's intensity, yet danger still lingers due to the lingering high water levels."
] |
1556659852054634496
|
The video depicts a nighttime scene of a severe flood in an urban area. The event unfolds near a building with illuminated signage, where a metal railing overlooks a channel of rapidly flowing water. Streetlights illuminate the dark scene, revealing the rising and then slightly receding water levels, which at their peak partially submerge the railing and nearby vegetation. The floodwaters create a turbulent and dangerous situation, with reduced visibility due to heavy rainfall and mist. As the video progresses, the intensity of the flood appears to decrease slightly, although the area remains heavily flooded and hazardous. The overall context suggests a sudden and intense flooding event that poses a threat to the surrounding environment and potential occupants.
|
On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.
|
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022",
"Cars and vehicles submerged in floodwater during the Seoul floods in Gangnam District, South Korea on 8 August 2022",
"Emergency responders rescuing individuals from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and property damage from floodwaters in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing out flooded homes and businesses in Seoul's Gangnam District after floods on 8 August 2022"
] |
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022 during floods",
"Cars and buses submerged in floodwater during the 8 August 2022 floods in Seoul's Gangnam District, South Korea",
"Emergency responders rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing debris and damaged belongings from their homes after floods in Gangnam District, Seoul, South Korea on 8 August 2022",
"Floodwaters overflowing from rivers and streams in Gangnam District, Seoul, South Korea during the floods on 8 August 2022"
] |
[
"People wading through floodwaters in the streets of Gangnam District Seoul South Korea during the floods on 8 August 2022",
"Cars and buses submerged in water on flooded roads in Gangnam District, Seoul, South Korea during the floods on August 8, 2022",
"Residents evacuating their homes and seeking higher ground in Seoul, South Korea, Gangnam District due to floods on 8 August 2022",
"Emergency responders, such as firefighters and police, rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and damaged furniture floating in the floodwaters during the 8 August 2022 floods in Gangnam District, Seoul, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2022%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A4%91%EB%B6%80_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District, it makes sense to extract events that are commonly associated with flooding in urban areas, such as people evacuating, floodwaters rising, and damage to infrastructure. These events are likely to be visualized in a video capturing the aftermath of the flood.\n\nEVENTS:\n1. People wading through floodwaters in the streets of Gangnam District\n2. Cars and buses submerged in water on flooded roads\n3. Residents evacuating their homes and seeking higher ground\n4. Emergency responders, such as firefighters and police, rescuing people from flooded areas\n5. Debris and damaged furniture floating in the floodwaters",
"en_article_excerpt": "On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.",
"en_article_url": "https://en.wikipedia.org/wiki/2022_South_Korean_floods",
"event_category": "disasters",
"event_name": "2022_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, particularly the Gangnam District and surrounding areas, it makes sense to extract events that are directly related to flooding and its immediate consequences, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visually captured in a video documenting the floods and their impact on the affected areas.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and vehicles submerged in floodwater\n3. Emergency responders rescuing individuals from flooded areas\n4. Debris and property damage from floodwaters\n5. Residents clearing out flooded homes and businesses",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District and surrounding areas, it makes sense to extract events that are commonly associated with flooding, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visualized in a video capturing the aftermath and response to the flooding in Seoul.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and buses submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Residents clearing debris and damaged belongings from their homes\n5. Floodwaters overflowing from rivers and streams in the affected areas",
"video_URL": "http://twitter.com/i/status/1556995436052873218",
"video_description": "Heavy rain in Seoul floods train stations, submerges vehicles. #Seoul #flooding #floods #SouthKorea",
"video_id": "1556995436052873218"
}
| 4
|
[
"The image shows a flooded stairway with water cascading down in a large volume, likely due to a flood or overflow. The walls and handrails are visible, and the water appears to be flowing rapidly. The stairs are covered with water, and the floor at the bottom is also submerged.",
"The water continues to rush down the stairway, indicating the flooding is still in progress. The intensity of the flow has not decreased, with water vigorously cascading over the steps. The handrails and tiled walls are still visible, and the entire area remains inundated.",
"The water continues to rush down the stairway, maintaining its vigorous cascade over the steps. The handrails and tiled walls remain visible, emphasizing the inundation of the area. The water appears slightly more turbulent, suggesting an increase in the intensity of the flooding.",
"The water is cascading down the stairs in a continuous, vigorous flow, now reaching a higher intensity as it forcefully rushes over the steps. The handrails and tiled walls are partially submerged, emphasizing the severity of the flooding. The water appears more turbulent, reflecting the increased force of the rush."
] |
1556995436052873218
|
A flooded stairway is depicted, with water rapidly cascading down the stairs due to a flood or overflow. The flooding continues throughout, with the water flow remaining vigorous and intense, eventually reaching a higher intensity and submerging the handrails and tiled walls, indicating a severe inundation of the area.
|
On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.
|
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022",
"Cars and vehicles submerged in floodwater during the Seoul floods in Gangnam District, South Korea on 8 August 2022",
"Emergency responders rescuing individuals from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and property damage from floodwaters in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing out flooded homes and businesses in Seoul's Gangnam District after floods on 8 August 2022"
] |
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022 during floods",
"Cars and buses submerged in floodwater during the 8 August 2022 floods in Seoul's Gangnam District, South Korea",
"Emergency responders rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing debris and damaged belongings from their homes after floods in Gangnam District, Seoul, South Korea on 8 August 2022",
"Floodwaters overflowing from rivers and streams in Gangnam District, Seoul, South Korea during the floods on 8 August 2022"
] |
[
"People wading through floodwaters in the streets of Gangnam District Seoul South Korea during the floods on 8 August 2022",
"Cars and buses submerged in water on flooded roads in Gangnam District, Seoul, South Korea during the floods on August 8, 2022",
"Residents evacuating their homes and seeking higher ground in Seoul, South Korea, Gangnam District due to floods on 8 August 2022",
"Emergency responders, such as firefighters and police, rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and damaged furniture floating in the floodwaters during the 8 August 2022 floods in Gangnam District, Seoul, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2022%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A4%91%EB%B6%80_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District, it makes sense to extract events that are commonly associated with flooding in urban areas, such as people evacuating, floodwaters rising, and damage to infrastructure. These events are likely to be visualized in a video capturing the aftermath of the flood.\n\nEVENTS:\n1. People wading through floodwaters in the streets of Gangnam District\n2. Cars and buses submerged in water on flooded roads\n3. Residents evacuating their homes and seeking higher ground\n4. Emergency responders, such as firefighters and police, rescuing people from flooded areas\n5. Debris and damaged furniture floating in the floodwaters",
"en_article_excerpt": "On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.",
"en_article_url": "https://en.wikipedia.org/wiki/2022_South_Korean_floods",
"event_category": "disasters",
"event_name": "2022_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, particularly the Gangnam District and surrounding areas, it makes sense to extract events that are directly related to flooding and its immediate consequences, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visually captured in a video documenting the floods and their impact on the affected areas.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and vehicles submerged in floodwater\n3. Emergency responders rescuing individuals from flooded areas\n4. Debris and property damage from floodwaters\n5. Residents clearing out flooded homes and businesses",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District and surrounding areas, it makes sense to extract events that are commonly associated with flooding, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visualized in a video capturing the aftermath and response to the flooding in Seoul.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and buses submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Residents clearing debris and damaged belongings from their homes\n5. Floodwaters overflowing from rivers and streams in the affected areas",
"video_URL": "http://twitter.com/i/status/1556833205671456768",
"video_description": "After heavy rain and #flashflood hit South Korea, water started coming inside the apartment as well. #Gangnam #Seoul #flooding #SouthKorea",
"video_id": "1556833205671456768"
}
| 4
|
[
"The image shows an elevator area with a wall that has a marble-like texture. There is a sign labeled \"B2\" indicating the basement level. The Korean text at the bottom reads \"서초아크로비스타,\" which suggests the name of the building or location. There are elevator buttons and a notice board on the wall, and an elevator door is partially visible on the right.",
"The image depicts the entrance to an elevator with the door closed. The surrounding walls still have the marble-like texture, and the sign above the door displays \"B2,\" confirming it is still at the basement level. The Korean text at the bottom remains \"서초아크로비스타,\" indicating the same building or location. The focus is now solely on the elevator door, suggesting someone is about to use it.",
"The current frame shows the elevator door closing, indicating that someone has just entered the elevator. The surrounding walls still have the marble-like texture, and the sign above the door still displays \"B2,\" confirming it is still at the basement level. The Korean text at the bottom remains \"서초아크로비스타,\" indicating the same building or location. The focus is now solely on the closing elevator door, suggesting the elevator is about to depart.",
"The current frame shows the elevator door completely closed, indicating that the elevator has likely started moving. The surrounding walls still have the marble-like texture, and the sign above the door still displays \"B2,\" confirming it is still at the basement level. The Korean text at the bottom remains \"서초아크로비스타,\" indicating the same building or location. The focus is now solely on the closed elevator door, suggesting the elevator is in transit."
] |
1556833205671456768
|
The video depicts a person entering an elevator at the basement level (B2) of a building named "서초아크로비스타" (Seocho Acrovista). The scene shows the elevator area, the entrance to the elevator, and the door closing as the person steps inside. The elevator then departs, indicated by the completely closed door, suggesting that the person is traveling to a different floor within the building. The entire sequence takes place in the elevator area of the basement level, with the building's name consistently displayed at the bottom of the frames.
|
On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.
|
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022",
"Cars and vehicles submerged in floodwater during the Seoul floods in Gangnam District, South Korea on 8 August 2022",
"Emergency responders rescuing individuals from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and property damage from floodwaters in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing out flooded homes and businesses in Seoul's Gangnam District after floods on 8 August 2022"
] |
[
"People evacuating flooded buildings and streets in Gangnam District, Seoul, South Korea on August 8, 2022 during floods",
"Cars and buses submerged in floodwater during the 8 August 2022 floods in Seoul's Gangnam District, South Korea",
"Emergency responders rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Residents clearing debris and damaged belongings from their homes after floods in Gangnam District, Seoul, South Korea on 8 August 2022",
"Floodwaters overflowing from rivers and streams in Gangnam District, Seoul, South Korea during the floods on 8 August 2022"
] |
[
"People wading through floodwaters in the streets of Gangnam District Seoul South Korea during the floods on 8 August 2022",
"Cars and buses submerged in water on flooded roads in Gangnam District, Seoul, South Korea during the floods on August 8, 2022",
"Residents evacuating their homes and seeking higher ground in Seoul, South Korea, Gangnam District due to floods on 8 August 2022",
"Emergency responders, such as firefighters and police, rescuing people from flooded areas in Gangnam District, Seoul, South Korea during the floods on 8 August 2022",
"Debris and damaged furniture floating in the floodwaters during the 8 August 2022 floods in Gangnam District, Seoul, South Korea"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2022%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A4%91%EB%B6%80_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District, it makes sense to extract events that are commonly associated with flooding in urban areas, such as people evacuating, floodwaters rising, and damage to infrastructure. These events are likely to be visualized in a video capturing the aftermath of the flood.\n\nEVENTS:\n1. People wading through floodwaters in the streets of Gangnam District\n2. Cars and buses submerged in water on flooded roads\n3. Residents evacuating their homes and seeking higher ground\n4. Emergency responders, such as firefighters and police, rescuing people from flooded areas\n5. Debris and damaged furniture floating in the floodwaters",
"en_article_excerpt": "On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.",
"en_article_url": "https://en.wikipedia.org/wiki/2022_South_Korean_floods",
"event_category": "disasters",
"event_name": "2022_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, particularly the Gangnam District and surrounding areas, it makes sense to extract events that are directly related to flooding and its immediate consequences, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visually captured in a video documenting the floods and their impact on the affected areas.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and vehicles submerged in floodwater\n3. Emergency responders rescuing individuals from flooded areas\n4. Debris and property damage from floodwaters\n5. Residents clearing out flooded homes and businesses",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 8 August 2022, large scale floods hit the capital of South Korea, Seoul, especially Gangnam District and surrounding areas.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about large-scale floods hitting Seoul, especially the Gangnam District and surrounding areas, it makes sense to extract events that are commonly associated with flooding, such as evacuation efforts, property damage, and rescue operations. These events are likely to be visualized in a video capturing the aftermath and response to the flooding in Seoul.\n\nEVENTS:\n1. People evacuating flooded buildings and streets\n2. Cars and buses submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Residents clearing debris and damaged belongings from their homes\n5. Floodwaters overflowing from rivers and streams in the affected areas",
"video_URL": "http://twitter.com/i/status/1557013913216352258",
"video_description": "Summers in South Korea be like... #southkorea #Monsoon2022 #monsoonseason #summer2022 #flooding #summersouthkorea #여름 #침수 #summertime",
"video_id": "1557013913216352258"
}
| 4
|
[
"A person is walking past a white car on a wet road, possibly in a city setting with some buildings and signs visible in the background.",
"The scene continues in a city setting where a significant amount of water has flooded the street, reflecting the lights from nearby buildings and signs. The reflection of neon lights suggests that the time is either dusk or nighttime. The water appears to be moving, indicating a possible current or rain. A few people are visible in the background, some holding umbrellas, suggesting that it is still raining. The overall atmosphere conveys a sense of urgency or caution due to the flooding.",
"The scene remains in the city setting with the street still significantly flooded, reflecting the neon lights from nearby buildings and signs, indicating it is still dusk or nighttime. The water continues to flow, suggesting ongoing rain or a persistent current. The scene is less crowded compared to the previous frames, with fewer people visible, some still holding umbrellas. The atmosphere retains a sense of caution, with the focus still on the flooding situation.",
"The scene continues to depict a city street during a flood, with the water level remaining high, reflecting the lights from surrounding buildings and signs. The water appears to be flowing steadily, indicating either ongoing rain or a sustained current. There are fewer people visible in the frame, and the overall atmosphere remains cautious, focusing on the impact of the flooding on the urban environment."
] |
1557013913216352258
|
A person walks through a flooded city street at dusk or nighttime, surrounded by reflections of neon lights from buildings and signs, indicating heavy rain. The flooding causes a sense of urgency and caution, with people in the background holding umbrellas and navigating the wet conditions. As the scene progresses, the street remains flooded with flowing water, reflecting the ongoing rain or current, and the atmosphere remains cautious, with a decreasing number of people visible, highlighting the impact of the flood on the urban environment.
|
A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.
|
[
"Flames engulfing the construction site in Icheon, South Korea on April 29, 2020, during a fire incident",
"Firefighters battling the blaze with water hoses during the Icheon, South Korea fire on 29 April 2020",
"Emergency responders rushing to rescue trapped workers during the Icheon, South Korea fire on April 29, 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Aerial footage of the construction site fire in Icheon, South Korea showing the extent of the damage on 29 April 2020"
] |
[
"Firefighters battling the blaze at the construction site in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing people trapped in the burning building during the Icheon, South Korea fire on 29 April 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Police investigating the cause of the fire in Icheon, South Korea on 29 April 2020",
"Rescue teams searching for missing people in the rubble after the Icheon, South Korea fire on 29 April 2020"
] |
[
"Flames engulfing the construction site in Icheon, South Korea during the fire on 29 April 2020",
"People running away from the burning building during the Icheon fire in South Korea on April 29, 2020",
"Firefighters arriving at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing injured individuals from a fire in Icheon, South Korea on 29 April 2020",
"Smoke billowing into the air as the fire rages on in Icheon, South Korea on 29 April 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%EC%9D%B4%EC%B2%9C_%EB%AC%BC%EB%A5%98%EC%84%BC%ED%84%B0_%EA%B3%B5%EC%82%AC%EC%9E%A5_%ED%99%94%EC%9E%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, as these would be the most visual and relevant aspects of the incident. The extracted events should focus on the fire, rescue efforts, and potential damage, as these are the most likely to be captured on video.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. People running away from the burning building\n3. Firefighters arriving at the scene\n4. Emergency responders rescuing injured individuals\n5. Smoke billowing into the air as the fire rages on",
"en_article_excerpt": "A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Icheon_fire",
"event_category": "disasters",
"event_name": "icheon_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the typical scenes that could be visualized in a video associated with a major fire at a construction site, such as the fire itself, rescue efforts, and the aftermath. Given the location and context of the event, it is likely that the extracted events will involve emergency responders, construction equipment, and damaged buildings.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. Firefighters battling the blaze with water hoses\n3. Emergency responders rushing to rescue trapped workers\n4. Paramedics treating injured people at the scene\n5. Aerial footage of the construction site showing the extent of the damage",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage assessment, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the construction site\n2. Emergency responders rescuing people trapped in the burning building\n3. Paramedics treating injured people at the scene\n4. Police investigating the cause of the fire\n5. Rescue teams searching for missing people in the rubble",
"video_URL": "http://youtube.com/watch?v=SCkWqQ49yUQ",
"video_description": "이천 물류센터 화재와 관련해 관할 지자체가 재난안전대책본부를 구성했습니다. 경기 이천시는 엄태준 시장을 본부장으로 30명 규모의 재난안전대책본부를 편성했다고 밝혔습니다. 의료구호반과 장비지원반 등 11개 반으로 꾸려진 대책본부는 환자 이송과 현장 수색 작업 등에 참여합니다. 신준명[shinjm7529@ytn.co.kr] ※ '당신의 제보가 뉴스가 됩니다' YTN은 여러분의 소중한 제보를 기다립니다. [카카오톡] YTN을 검색해 채널 추가 [전화] 02-398-8585 [메일] social@ytn.co.kr [온라인 제보] www.ytn.co.kr ▶ 기사 원문 : https://www.ytn.co.kr/_ln/0103_202004292044009865 ▶ 제보 안내 : http://goo.gl/gEvsAL, 모바일앱, social@ytn.co.kr, #2424 ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN & YTN PLUS 무단 전재 및 재배포금지",
"video_id": "SCkWqQ49yUQ"
}
| 4
|
[
"The image shows a severely damaged building with significant fire damage evident on the exterior walls and windows. In front of the building, there are several barrels and a blue tent set up. The text in the image indicates that this is a warehouse fire involving a waste recycling center. The scene suggests a post-fire investigation or assessment, with emergency or response personnel likely present.",
"The image shows a close-up view of a damaged building, likely the same one from the previous frame. The exterior wall and window are heavily charred and blackened, indicating extensive fire damage. The text in the image suggests that this is a warehouse fire involving a waste recycling center. The scene continues to suggest a post-fire investigation or assessment, with emergency or response personnel likely present.",
"The image continues to focus on the exterior of a damaged building, consistent with the previous frame. The walls show significant scorch marks and discoloration from fire damage, with visible blackened areas and charred textures. The text overlay suggests that the building, likely a warehouse, experienced a major fire at a waste recycling center. The scene continues to indicate a post-disaster assessment, with emergency personnel likely present, assessing the extent of the damage and possibly investigating the cause of the fire.",
"The focus remains on the exterior of the damaged building, which continues to display significant scorch marks and discoloration from fire damage. The walls show visible blackened areas and charred textures, consistent with the aftermath of a major fire. The text overlay indicates that the building is a warehouse at a waste recycling center, which has experienced a significant blaze. The scene suggests ongoing post-disaster assessment, with emergency personnel likely present to assess the extent of the damage and investigate the cause of the fire. The image emphasizes the severity of the fire and the resulting damage to the structure."
] |
SCkWqQ49yUQ
|
A severe fire occurred at a waste recycling center, causing significant damage to a warehouse building. The fire resulted in extensive scorch marks, discoloration, and charred textures on the exterior walls and windows. Emergency or response personnel are present, conducting a post-fire investigation and assessment to determine the extent of the damage and likely investigating the cause of the fire. The scene shows a blue tent and barrels set up in front of the building, indicating a thorough examination of the area. The video depicts the aftermath of the fire, highlighting the severity of the damage to the warehouse structure.
|
A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.
|
[
"Flames engulfing the construction site in Icheon, South Korea on April 29, 2020, during a fire incident",
"Firefighters battling the blaze with water hoses during the Icheon, South Korea fire on 29 April 2020",
"Emergency responders rushing to rescue trapped workers during the Icheon, South Korea fire on April 29, 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Aerial footage of the construction site fire in Icheon, South Korea showing the extent of the damage on 29 April 2020"
] |
[
"Firefighters battling the blaze at the construction site in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing people trapped in the burning building during the Icheon, South Korea fire on 29 April 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Police investigating the cause of the fire in Icheon, South Korea on 29 April 2020",
"Rescue teams searching for missing people in the rubble after the Icheon, South Korea fire on 29 April 2020"
] |
[
"Flames engulfing the construction site in Icheon, South Korea during the fire on 29 April 2020",
"People running away from the burning building during the Icheon fire in South Korea on April 29, 2020",
"Firefighters arriving at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing injured individuals from a fire in Icheon, South Korea on 29 April 2020",
"Smoke billowing into the air as the fire rages on in Icheon, South Korea on 29 April 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%EC%9D%B4%EC%B2%9C_%EB%AC%BC%EB%A5%98%EC%84%BC%ED%84%B0_%EA%B3%B5%EC%82%AC%EC%9E%A5_%ED%99%94%EC%9E%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, as these would be the most visual and relevant aspects of the incident. The extracted events should focus on the fire, rescue efforts, and potential damage, as these are the most likely to be captured on video.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. People running away from the burning building\n3. Firefighters arriving at the scene\n4. Emergency responders rescuing injured individuals\n5. Smoke billowing into the air as the fire rages on",
"en_article_excerpt": "A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Icheon_fire",
"event_category": "disasters",
"event_name": "icheon_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the typical scenes that could be visualized in a video associated with a major fire at a construction site, such as the fire itself, rescue efforts, and the aftermath. Given the location and context of the event, it is likely that the extracted events will involve emergency responders, construction equipment, and damaged buildings.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. Firefighters battling the blaze with water hoses\n3. Emergency responders rushing to rescue trapped workers\n4. Paramedics treating injured people at the scene\n5. Aerial footage of the construction site showing the extent of the damage",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage assessment, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the construction site\n2. Emergency responders rescuing people trapped in the burning building\n3. Paramedics treating injured people at the scene\n4. Police investigating the cause of the fire\n5. Rescue teams searching for missing people in the rubble",
"video_URL": "http://youtube.com/watch?v=oRy1PP3KzDs",
"video_description": "29일 오후 경기도 이천 물류창고 공사현장에서 화재가 발생해 큰 인명피해를 냈습니다. 오후 7시30분 현재 확인된 사망자만 36명입니다.",
"video_id": "oRy1PP3KzDs"
}
| 4
|
[
"The video shows a large building engulfed in smoke, indicating a fire incident. Firefighters are seen spraying water on the building to control the blaze. The text in Korean mentions a urethane foam factory and a chemical explosion.",
"Firefighters are actively combating the blaze by spraying water on a building that is heavily emitting smoke. The Korean text confirms that there are 11 casualties and 14 people are in critical condition, consistent with the ongoing fire and explosion incident at a urethane foam factory.",
"Firefighters are still actively combating the fire by spraying water on a building that is heavily emitting smoke. The Korean text at the top of the image indicates that a person was found alive amidst the fire and explosion incident at a urethane foam factory, suggesting a hopeful development in the rescue efforts.",
"The current frame displays a graphic with the logo of Korea's National Emergency Management Agency (KEMAN) alongside the Korean text '영상제공: 경기도소방재난본부' which translates to 'Video provided by: Gyeonggi-do Fire and Disaster Prevention Headquarters.' This indicates that the footage being viewed was provided by the local fire and disaster prevention authorities in Gyeonggi-do, which is part of the national response to the incident. The graphic serves as a transition or credit screen, acknowledging the source of the footage shown in the previous frames."
] |
oRy1PP3KzDs
|
A fire and chemical explosion occurred at a urethane foam factory, resulting in a large blaze with significant smoke emission. Firefighters responded to the incident, actively spraying water to control the fire. The explosion caused 11 casualties and left 14 people in critical condition. Despite the severity of the situation, rescue efforts yielded a positive outcome with at least one person being found alive amidst the disaster. The footage of the incident was provided by the Gyeonggi-do Fire and Disaster Prevention Headquarters, indicating an official response and documentation of the event.
|
A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.
|
[
"Flames engulfing the construction site in Icheon, South Korea on April 29, 2020, during a fire incident",
"Firefighters battling the blaze with water hoses during the Icheon, South Korea fire on 29 April 2020",
"Emergency responders rushing to rescue trapped workers during the Icheon, South Korea fire on April 29, 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Aerial footage of the construction site fire in Icheon, South Korea showing the extent of the damage on 29 April 2020"
] |
[
"Firefighters battling the blaze at the construction site in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing people trapped in the burning building during the Icheon, South Korea fire on 29 April 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Police investigating the cause of the fire in Icheon, South Korea on 29 April 2020",
"Rescue teams searching for missing people in the rubble after the Icheon, South Korea fire on 29 April 2020"
] |
[
"Flames engulfing the construction site in Icheon, South Korea during the fire on 29 April 2020",
"People running away from the burning building during the Icheon fire in South Korea on April 29, 2020",
"Firefighters arriving at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing injured individuals from a fire in Icheon, South Korea on 29 April 2020",
"Smoke billowing into the air as the fire rages on in Icheon, South Korea on 29 April 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%EC%9D%B4%EC%B2%9C_%EB%AC%BC%EB%A5%98%EC%84%BC%ED%84%B0_%EA%B3%B5%EC%82%AC%EC%9E%A5_%ED%99%94%EC%9E%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, as these would be the most visual and relevant aspects of the incident. The extracted events should focus on the fire, rescue efforts, and potential damage, as these are the most likely to be captured on video.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. People running away from the burning building\n3. Firefighters arriving at the scene\n4. Emergency responders rescuing injured individuals\n5. Smoke billowing into the air as the fire rages on",
"en_article_excerpt": "A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Icheon_fire",
"event_category": "disasters",
"event_name": "icheon_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the typical scenes that could be visualized in a video associated with a major fire at a construction site, such as the fire itself, rescue efforts, and the aftermath. Given the location and context of the event, it is likely that the extracted events will involve emergency responders, construction equipment, and damaged buildings.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. Firefighters battling the blaze with water hoses\n3. Emergency responders rushing to rescue trapped workers\n4. Paramedics treating injured people at the scene\n5. Aerial footage of the construction site showing the extent of the damage",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage assessment, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the construction site\n2. Emergency responders rescuing people trapped in the burning building\n3. Paramedics treating injured people at the scene\n4. Police investigating the cause of the fire\n5. Rescue teams searching for missing people in the rubble",
"video_URL": "http://youtube.com/watch?v=rqRqdg-oxkU",
"video_description": "이번에도 샌드위치 패널... 처참한 이천 물류센터 화재현장",
"video_id": "rqRqdg-oxkU"
}
| 4
|
[
"The image shows a large building engulfed in thick black smoke, indicating a significant fire. Water jets are being directed at the building from the ground, suggesting firefighting efforts are underway.",
"The firefighting efforts continue as water jets continue to be directed at the building, which is now emitting thick black smoke from multiple points, indicating the fire is still burning intensely inside.",
"The firefighting efforts persist, with the building continuing to emit thick black smoke from multiple points. Water jets remain directed at the structure, attempting to control the fire, which appears to be still burning intensely inside.",
"The firefighting efforts persist, with the building continuing to emit thick black smoke from multiple points. The firefighters are actively engaged in spraying water onto the structure, attempting to control the fire, which still appears to be burning intensely inside."
] |
rqRqdg-oxkU
|
A large building is on fire, with thick black smoke emanating from multiple points, indicating a significant and intense fire. Firefighters are actively engaged in combating the blaze, using water jets to try and control the fire, which continues to burn intensely inside the building throughout the duration of the video. The firefighting efforts are persistent, with the firefighters continually spraying water onto the structure in an attempt to extinguish the fire.
|
A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.
|
[
"Flames engulfing the construction site in Icheon, South Korea on April 29, 2020, during a fire incident",
"Firefighters battling the blaze with water hoses during the Icheon, South Korea fire on 29 April 2020",
"Emergency responders rushing to rescue trapped workers during the Icheon, South Korea fire on April 29, 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Aerial footage of the construction site fire in Icheon, South Korea showing the extent of the damage on 29 April 2020"
] |
[
"Firefighters battling the blaze at the construction site in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing people trapped in the burning building during the Icheon, South Korea fire on 29 April 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Police investigating the cause of the fire in Icheon, South Korea on 29 April 2020",
"Rescue teams searching for missing people in the rubble after the Icheon, South Korea fire on 29 April 2020"
] |
[
"Flames engulfing the construction site in Icheon, South Korea during the fire on 29 April 2020",
"People running away from the burning building during the Icheon fire in South Korea on April 29, 2020",
"Firefighters arriving at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing injured individuals from a fire in Icheon, South Korea on 29 April 2020",
"Smoke billowing into the air as the fire rages on in Icheon, South Korea on 29 April 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%EC%9D%B4%EC%B2%9C_%EB%AC%BC%EB%A5%98%EC%84%BC%ED%84%B0_%EA%B3%B5%EC%82%AC%EC%9E%A5_%ED%99%94%EC%9E%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, as these would be the most visual and relevant aspects of the incident. The extracted events should focus on the fire, rescue efforts, and potential damage, as these are the most likely to be captured on video.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. People running away from the burning building\n3. Firefighters arriving at the scene\n4. Emergency responders rescuing injured individuals\n5. Smoke billowing into the air as the fire rages on",
"en_article_excerpt": "A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Icheon_fire",
"event_category": "disasters",
"event_name": "icheon_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the typical scenes that could be visualized in a video associated with a major fire at a construction site, such as the fire itself, rescue efforts, and the aftermath. Given the location and context of the event, it is likely that the extracted events will involve emergency responders, construction equipment, and damaged buildings.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. Firefighters battling the blaze with water hoses\n3. Emergency responders rushing to rescue trapped workers\n4. Paramedics treating injured people at the scene\n5. Aerial footage of the construction site showing the extent of the damage",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage assessment, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the construction site\n2. Emergency responders rescuing people trapped in the burning building\n3. Paramedics treating injured people at the scene\n4. Police investigating the cause of the fire\n5. Rescue teams searching for missing people in the rubble",
"video_URL": "http://youtube.com/watch?v=6Z2tI74NGGQ",
"video_description": "[당시 현장 목격자 : 고개 돌리는 순간 연기가 엄청 심하게 나오더라고요. 설마 저게 불일까 그랬는데, 폭발음과 함께 불꽃이 일기 시작한 거죠.] 삽시간에 불길 번져 인명 피해 잇따라 \"우레탄 작업 중 폭발\" 추정 [서승현 / 경기 이천소방서장 : 화상 당하신 분들이 옷이 다 탄 걸로 봐서는 유증기가 폭발하면서 일어난 것으로 추정….] 완공 2개월 앞두고 참사 또 발생한 대형 화재 사상자 속출 ▶ 기사 원문 : https://www.ytn.co.kr/_ln/0103_202004292050572443 ▶ 제보 안내 : http://goo.gl/gEvsAL, 모바일앱, social@ytn.co.kr, #2424 ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN & YTN PLUS 무단 전재 및 재배포금지",
"video_id": "6Z2tI74NGGQ"
}
| 4
|
[
"The current frame depicts a large fire engulfing a structure, with thick black smoke billowing into the air. A fire truck is seen spraying water towards the flames. There are a few people visible in the foreground, likely firefighters or bystanders. The text in Korean at the bottom reads \"당시 현장 목격자 설명부터 불이 나는데,\" which translates to \"At the time, a witness at the scene described it as a fire starting.\"",
"The fire continues to burn intensely with thick smoke rising from the building. The fire truck is still actively spraying water onto the fire. A few people remain in the foreground, likely involved in firefighting efforts or observing the scene.",
"Firefighters in full gear are actively engaged near a fire truck. One firefighter is bent over, handling equipment or a hose, while others stand by. The fire truck's compartments are open, revealing firefighting tools and supplies, indicating ongoing efforts to combat the fire.",
"Firefighters in full gear are entering a building through a damaged entrance. They are equipped with oxygen tanks and face masks, suggesting they are preparing to navigate potentially hazardous conditions inside the structure. The building's exterior shows visible fire damage, including charred and cracked materials. The text in the image indicates that a large fire has resulted in the death of the property owner, highlighting the severity and impact of the incident."
] |
6Z2tI74NGGQ
|
A severe fire engulfs a structure, prompting a response from firefighters who arrive in a fire truck to combat the blaze. The fire burns intensely, producing thick black smoke, and firefighters work to extinguish it, using water and equipment from the truck. As the situation unfolds, firefighters, equipped with protective gear, enter the damaged building to navigate its interior, indicating a serious and potentially hazardous situation. The fire results in significant damage to the property and tragically claims the life of the property owner, underscoring the gravity of the incident.
|
A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.
|
[
"Flames engulfing the construction site in Icheon, South Korea on April 29, 2020, during a fire incident",
"Firefighters battling the blaze with water hoses during the Icheon, South Korea fire on 29 April 2020",
"Emergency responders rushing to rescue trapped workers during the Icheon, South Korea fire on April 29, 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Aerial footage of the construction site fire in Icheon, South Korea showing the extent of the damage on 29 April 2020"
] |
[
"Firefighters battling the blaze at the construction site in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing people trapped in the burning building during the Icheon, South Korea fire on 29 April 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Police investigating the cause of the fire in Icheon, South Korea on 29 April 2020",
"Rescue teams searching for missing people in the rubble after the Icheon, South Korea fire on 29 April 2020"
] |
[
"Flames engulfing the construction site in Icheon, South Korea during the fire on 29 April 2020",
"People running away from the burning building during the Icheon fire in South Korea on April 29, 2020",
"Firefighters arriving at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing injured individuals from a fire in Icheon, South Korea on 29 April 2020",
"Smoke billowing into the air as the fire rages on in Icheon, South Korea on 29 April 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%EC%9D%B4%EC%B2%9C_%EB%AC%BC%EB%A5%98%EC%84%BC%ED%84%B0_%EA%B3%B5%EC%82%AC%EC%9E%A5_%ED%99%94%EC%9E%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, as these would be the most visual and relevant aspects of the incident. The extracted events should focus on the fire, rescue efforts, and potential damage, as these are the most likely to be captured on video.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. People running away from the burning building\n3. Firefighters arriving at the scene\n4. Emergency responders rescuing injured individuals\n5. Smoke billowing into the air as the fire rages on",
"en_article_excerpt": "A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Icheon_fire",
"event_category": "disasters",
"event_name": "icheon_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the typical scenes that could be visualized in a video associated with a major fire at a construction site, such as the fire itself, rescue efforts, and the aftermath. Given the location and context of the event, it is likely that the extracted events will involve emergency responders, construction equipment, and damaged buildings.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. Firefighters battling the blaze with water hoses\n3. Emergency responders rushing to rescue trapped workers\n4. Paramedics treating injured people at the scene\n5. Aerial footage of the construction site showing the extent of the damage",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage assessment, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the construction site\n2. Emergency responders rescuing people trapped in the burning building\n3. Paramedics treating injured people at the scene\n4. Police investigating the cause of the fire\n5. Rescue teams searching for missing people in the rubble",
"video_URL": "http://youtube.com/watch?v=JCKK0gu12zI",
"video_description": "경기 이천시 물류센터 화재가 대응 2단계로 격상한 가운데 인명피해가 속출하고 있다. 29일 경기도소방재난본부에 따르면 이날 오후 1시 32분 이천시 모가면 소고리 연면적 약 1만1043㎡인 물류센터 신축 공사현장에서 화재가 발생해 11명이 인명피해를 입었다.",
"video_id": "JCKK0gu12zI"
}
| 4
|
[
"The image shows a large building engulfed in thick smoke, indicating a significant fire. Firefighters can be seen actively spraying water onto the burning structure in an attempt to control and extinguish the flames. The text in the top left corner of the image is in Korean and reads \"이천 물류센터 화재 현장,\" which translates to \"Icheon Logistics Center Fire Scene.\" The logo of the broadcasting network \"뉴스핌\" is visible in the top right corner.",
"The image continues to depict the large building engulfed in thick smoke, indicating the ongoing significant fire. Firefighters are still actively spraying water onto the burning structure, with a particular focus on the area where smoke is billowing out. The text \"이천 물류센터 화재 현장\" is still present, emphasizing the location of the incident. The logo \"뉴스핌\" remains visible, maintaining the continuity of the broadcasting network.",
"The image still shows the large building engulfed in thick smoke, indicating the significant fire is ongoing. Firefighters are actively spraying water onto the burning structure, with a particular focus on the area where smoke is billowing out. The text \"이천 물류센터 화재 현장\" is still present, emphasizing the location of the incident. The logo \"뉴스핌\" remains visible, maintaining the continuity of the broadcasting network.",
"The current frame shows a logo with Korean text \"뉴스핌\" on a white background."
] |
JCKK0gu12zI
|
A significant fire broke out at the Icheon Logistics Center, with firefighters actively working to control and extinguish the flames by spraying water onto the burning structure. The fire produced thick smoke, and the responders focused their efforts on the areas with the most intense smoke. The incident was reported by 뉴스핌, a broadcasting network, which covered the event and provided a live scene of the fire.
|
A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.
|
[
"Flames engulfing the construction site in Icheon, South Korea on April 29, 2020, during a fire incident",
"Firefighters battling the blaze with water hoses during the Icheon, South Korea fire on 29 April 2020",
"Emergency responders rushing to rescue trapped workers during the Icheon, South Korea fire on April 29, 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Aerial footage of the construction site fire in Icheon, South Korea showing the extent of the damage on 29 April 2020"
] |
[
"Firefighters battling the blaze at the construction site in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing people trapped in the burning building during the Icheon, South Korea fire on 29 April 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Police investigating the cause of the fire in Icheon, South Korea on 29 April 2020",
"Rescue teams searching for missing people in the rubble after the Icheon, South Korea fire on 29 April 2020"
] |
[
"Flames engulfing the construction site in Icheon, South Korea during the fire on 29 April 2020",
"People running away from the burning building during the Icheon fire in South Korea on April 29, 2020",
"Firefighters arriving at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing injured individuals from a fire in Icheon, South Korea on 29 April 2020",
"Smoke billowing into the air as the fire rages on in Icheon, South Korea on 29 April 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%EC%9D%B4%EC%B2%9C_%EB%AC%BC%EB%A5%98%EC%84%BC%ED%84%B0_%EA%B3%B5%EC%82%AC%EC%9E%A5_%ED%99%94%EC%9E%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, as these would be the most visual and relevant aspects of the incident. The extracted events should focus on the fire, rescue efforts, and potential damage, as these are the most likely to be captured on video.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. People running away from the burning building\n3. Firefighters arriving at the scene\n4. Emergency responders rescuing injured individuals\n5. Smoke billowing into the air as the fire rages on",
"en_article_excerpt": "A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Icheon_fire",
"event_category": "disasters",
"event_name": "icheon_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the typical scenes that could be visualized in a video associated with a major fire at a construction site, such as the fire itself, rescue efforts, and the aftermath. Given the location and context of the event, it is likely that the extracted events will involve emergency responders, construction equipment, and damaged buildings.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. Firefighters battling the blaze with water hoses\n3. Emergency responders rushing to rescue trapped workers\n4. Paramedics treating injured people at the scene\n5. Aerial footage of the construction site showing the extent of the damage",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage assessment, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the construction site\n2. Emergency responders rescuing people trapped in the burning building\n3. Paramedics treating injured people at the scene\n4. Police investigating the cause of the fire\n5. Rescue teams searching for missing people in the rubble",
"video_URL": "http://youtube.com/watch?v=ZFWsFwpfGlE",
"video_description": "경찰 오늘 이천 물류창고 화재 중간 수사결과 발표 [전국네트워크] 경기남부지방경찰청이 오늘 오전 38명의 목숨을 앗아간 '경기도 이천 물류창고 화재' 사고에 대한 중간 수사 결과를 발표합니다. 지난 4월 29일 경기 이천시의 한 물류창고 공사현장에서 불이 나 38명이 숨지고 10명이 다친 바 있는데, 화재 원인은 지하 2층 우레탄 작업 중 발생한 미상의 화원에 의한 폭발로 추정되고 있습니다. 경찰은 4차례의 대규모 현장 감식과 함께 화재와 관련된 80여 명을 조사해 이 가운데 17명에 대해 업무상 과실치사상 혐의 등을 적용해 입건했습니다. ☞ MBN 유튜브 구독하기 ☞ https://goo.gl/6ZsJGT 📢 MBN 유튜브 커뮤니티https://www.youtube.com/user/mbn/community?disable_polymer=1 MBN 페이스북 http://www.facebook.com/mbntv MBN 인스타그램 https://www.instagram.com/mbn_news/ #MBN뉴스#전국네트워크#이천물류창고",
"video_id": "ZFWsFwpfGlE"
}
| 4
|
[
"The image shows a partially destroyed building, with visible damage and burn marks on its structure. The building appears to be severely damaged by fire. A text overlay in Korean reads: \"이천 물류창고 화재 중간 수사결과 발표,\" which translates to \"Icheon Logistics Warehouse Fire Intermediate Investigation Results Announcement.\" The logo of a news channel, MBN, is also visible in the top right corner.",
"In the current frame, the scene has shifted to a group of people in white protective suits walking along a road, with traffic cones and vehicles visible in the background. One of the individuals is wearing a gas mask and carrying a camera, possibly indicating a focus on safety and documentation in response to the fire incident. The text overlay in Korean reads: \"이천 물류창고 화재 중간 수사결과 발표,\" which translates to \"Icheon Logistics Warehouse Fire Intermediate Investigation Results Announcement.\" The presence of the news channel logo, MBN, confirms the continuity of the news coverage related to the fire incident.",
"The focus of the current frame is on the damaged structure, showing significant fire damage. The building appears heavily charred with windows missing and debris scattered around. Yellow caution tape is visible, indicating the area is off-limits. The text overlay in Korean reads: \"이천 물류창고 화재 중간 수사결과 발표,\" which translates to \"Icheon Logistics Warehouse Fire Intermediate Investigation Results Announcement.\" This frame suggests the aftermath of the fire, with investigators possibly examining the extent of the damage.",
"The current frame shows a group of individuals walking through a heavily damaged and debris-strewn area. The individuals appear to be investigators or officials, as indicated by their attire and the presence of a clipboard, suggesting they are examining the scene. The ground is littered with rubble and remnants of a fire, consistent with the aftermath of a significant fire incident. The presence of caution tape and the context from the previous frame indicate that this is an investigation site, likely related to the logistics warehouse fire mentioned earlier."
] |
ZFWsFwpfGlE
|
A news report by MBN covers the intermediate investigation results of a significant fire at a logistics warehouse in Icheon. The video depicts the severely damaged building with visible burn marks and destruction, indicating a substantial fire incident. Investigators, dressed in protective gear, are shown walking through the debris-strewn area, examining the scene and documenting the damage. The presence of caution tape and safety equipment suggests a focus on safety and a thorough investigation into the cause and extent of the fire. The report appears to be presenting the findings of the intermediate investigation, providing an update on the status of the inquiry into the logistics warehouse fire.
|
A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.
|
[
"Flames engulfing the construction site in Icheon, South Korea on April 29, 2020, during a fire incident",
"Firefighters battling the blaze with water hoses during the Icheon, South Korea fire on 29 April 2020",
"Emergency responders rushing to rescue trapped workers during the Icheon, South Korea fire on April 29, 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Aerial footage of the construction site fire in Icheon, South Korea showing the extent of the damage on 29 April 2020"
] |
[
"Firefighters battling the blaze at the construction site in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing people trapped in the burning building during the Icheon, South Korea fire on 29 April 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Police investigating the cause of the fire in Icheon, South Korea on 29 April 2020",
"Rescue teams searching for missing people in the rubble after the Icheon, South Korea fire on 29 April 2020"
] |
[
"Flames engulfing the construction site in Icheon, South Korea during the fire on 29 April 2020",
"People running away from the burning building during the Icheon fire in South Korea on April 29, 2020",
"Firefighters arriving at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing injured individuals from a fire in Icheon, South Korea on 29 April 2020",
"Smoke billowing into the air as the fire rages on in Icheon, South Korea on 29 April 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%EC%9D%B4%EC%B2%9C_%EB%AC%BC%EB%A5%98%EC%84%BC%ED%84%B0_%EA%B3%B5%EC%82%AC%EC%9E%A5_%ED%99%94%EC%9E%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, as these would be the most visual and relevant aspects of the incident. The extracted events should focus on the fire, rescue efforts, and potential damage, as these are the most likely to be captured on video.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. People running away from the burning building\n3. Firefighters arriving at the scene\n4. Emergency responders rescuing injured individuals\n5. Smoke billowing into the air as the fire rages on",
"en_article_excerpt": "A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Icheon_fire",
"event_category": "disasters",
"event_name": "icheon_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the typical scenes that could be visualized in a video associated with a major fire at a construction site, such as the fire itself, rescue efforts, and the aftermath. Given the location and context of the event, it is likely that the extracted events will involve emergency responders, construction equipment, and damaged buildings.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. Firefighters battling the blaze with water hoses\n3. Emergency responders rushing to rescue trapped workers\n4. Paramedics treating injured people at the scene\n5. Aerial footage of the construction site showing the extent of the damage",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage assessment, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the construction site\n2. Emergency responders rescuing people trapped in the burning building\n3. Paramedics treating injured people at the scene\n4. Police investigating the cause of the fire\n5. Rescue teams searching for missing people in the rubble",
"video_URL": "http://youtube.com/watch?v=2SCMvAXvBLM",
"video_description": "경기도 이천 모가면의 한 물류창고 공사현장에서 29일 오후 1시32분 쯤 불이 나 현재까지 5명이 사망한 것으로 확인되고 있습니다. 소방당국은 대응 2단계를 발령하고 소방대원 60명과 펌프차 등 장비 27대를 투입해 진화작업을 벌이고 있습니다. 소방 관계자에 따르면 물류창고의 지하 공사현장에서 불이 시작된 것으로 보이며, 사망자는 앞으로 더 늘어날 전망입니다. 소방 대응단계는 화재 규모에 따라 1~3단계로 나뉘어지는데, 1단계는 4개 이하 소방서가 합동 대응하며 2단계는 5~9개 소방서, 3단계는 10개 이상 소방서가 함께 진화작업을 벌입니다. https://imnews.imbc.com/original/todayfield/5749059_32167.html #경기도 #이천 #물류창고 #화재 #소방당국 #대응 2단계 #진화작업",
"video_id": "2SCMvAXvBLM"
}
| 4
|
[
"The image shows an aerial view of a large fire with thick black smoke billowing from a building complex. There are several buildings and vehicles visible in the vicinity, and emergency services appear to be present, indicated by the presence of vehicles and equipment near the buildings. The text \"Right Now\" is visible in the top right corner, suggesting this is a live or recent event being captured.",
"The image shows an aerial view of a large fire with thick black smoke billowing from a building complex. There are several buildings and vehicles visible in the vicinity, and emergency services appear to be present, indicated by the presence of vehicles and equipment near the buildings. The text \"Right Now\" is visible in the top right corner, suggesting this is a live or recent event being captured.",
"The current frame shows a continued aerial view of the large fire, with thick black smoke still billowing from the building complex. Several buildings and vehicles remain visible in the vicinity, and emergency services are still present, indicated by the presence of vehicles and equipment near the buildings. The text \"Right Now\" is still visible in the top right corner, suggesting this is a live or recent event being captured. There is no significant change in the smoke's direction or intensity, indicating the fire is still ongoing and emergency services are actively managing the situation.",
"The current frame shows a continued aerial view of the large fire, with thick black smoke still billowing from the building complex. Several buildings and vehicles remain visible in the vicinity, and emergency services are still present, indicated by the presence of vehicles and equipment near the buildings. The text \"Right Now\" is still visible in the top right corner, suggesting this is a live or recent event being captured. There is no significant change in the smoke's direction or intensity, indicating the fire is still ongoing and emergency services are actively managing the situation."
] |
2SCMvAXvBLM
|
A large fire is ongoing at a building complex, with thick black smoke billowing into the air. Emergency services are present and actively managing the situation, as indicated by their vehicles and equipment near the buildings. The event is being captured in real-time or recently, as suggested by the "Right Now" text in the corner of the frames. The fire continues unabated across the observed frames, with no significant change in smoke direction or intensity, indicating an ongoing emergency response effort.
|
A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.
|
[
"Flames engulfing the construction site in Icheon, South Korea on April 29, 2020, during a fire incident",
"Firefighters battling the blaze with water hoses during the Icheon, South Korea fire on 29 April 2020",
"Emergency responders rushing to rescue trapped workers during the Icheon, South Korea fire on April 29, 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Aerial footage of the construction site fire in Icheon, South Korea showing the extent of the damage on 29 April 2020"
] |
[
"Firefighters battling the blaze at the construction site in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing people trapped in the burning building during the Icheon, South Korea fire on 29 April 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Police investigating the cause of the fire in Icheon, South Korea on 29 April 2020",
"Rescue teams searching for missing people in the rubble after the Icheon, South Korea fire on 29 April 2020"
] |
[
"Flames engulfing the construction site in Icheon, South Korea during the fire on 29 April 2020",
"People running away from the burning building during the Icheon fire in South Korea on April 29, 2020",
"Firefighters arriving at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing injured individuals from a fire in Icheon, South Korea on 29 April 2020",
"Smoke billowing into the air as the fire rages on in Icheon, South Korea on 29 April 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%EC%9D%B4%EC%B2%9C_%EB%AC%BC%EB%A5%98%EC%84%BC%ED%84%B0_%EA%B3%B5%EC%82%AC%EC%9E%A5_%ED%99%94%EC%9E%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, as these would be the most visual and relevant aspects of the incident. The extracted events should focus on the fire, rescue efforts, and potential damage, as these are the most likely to be captured on video.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. People running away from the burning building\n3. Firefighters arriving at the scene\n4. Emergency responders rescuing injured individuals\n5. Smoke billowing into the air as the fire rages on",
"en_article_excerpt": "A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Icheon_fire",
"event_category": "disasters",
"event_name": "icheon_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the typical scenes that could be visualized in a video associated with a major fire at a construction site, such as the fire itself, rescue efforts, and the aftermath. Given the location and context of the event, it is likely that the extracted events will involve emergency responders, construction equipment, and damaged buildings.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. Firefighters battling the blaze with water hoses\n3. Emergency responders rushing to rescue trapped workers\n4. Paramedics treating injured people at the scene\n5. Aerial footage of the construction site showing the extent of the damage",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage assessment, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the construction site\n2. Emergency responders rescuing people trapped in the burning building\n3. Paramedics treating injured people at the scene\n4. Police investigating the cause of the fire\n5. Rescue teams searching for missing people in the rubble",
"video_URL": "http://youtube.com/watch?v=n8dcNr14bm8",
"video_description": "38명이 숨진 경기도 이천 물류창고 공사장 화재 현장에서 경찰이 아직 수습되지 않은 유해 일부와 유류품 등을 찾기 위한 정밀 수색을 하고 있습니다. 경기남부지방경찰청 과학수사대는 오늘(2일) 오전 9시부터 중장비와 과학수사요원 15명을 투입해 이천시 모가면 물류창고 공사 현장을 정밀 수색하고 있습니다. 오늘 수색은 사고로 숨진 사망자들의 유해 가운데 아직 수습되지 않은 일부와 유류품을 찾는 데 주력한다는 방침입니다. 앞서 경찰은 국립과학수사연구원에 의뢰한 DNA 조사를 통해 신원이 확인되지 않은 9명 가운데 8명의 신원을 확인해 유가족들에게 전했습니다. 경찰은 남은 신원 미확인자 1명의 DNA 검사 결과도 오늘 안에 나올 것으로 기대하고 있습니다. ※ '당신의 제보가 뉴스가 됩니다' YTN은 여러분의 소중한 제보를 기다립니다. [카카오톡] YTN을 검색해 채널 추가 [전화] 02-398-8585 [메일] social@ytn.co.kr [온라인 제보] www.ytn.co.kr ▶ 기사 원문 : https://www.ytn.co.kr/_ln/0103_202005021020478614 ▶ 제보 안내 : http://goo.gl/gEvsAL, 모바일앱, social@ytn.co.kr, #2424 ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN & YTN PLUS 무단 전재 및 재배포금지",
"video_id": "n8dcNr14bm8"
}
| 4
|
[
"The image shows a group of individuals in white protective suits and masks walking through an area with debris and rubble, indicating a site of destruction or an emergency situation. There is a yellow excavator in the background, and the surrounding area appears to be in disrepair. The text in Korean at the bottom of the image suggests that this is related to a demolition or accident investigation, mentioning \"경찰, 이천 물류창고 화재현장 정밀 수색\" which translates to \"Police, detailed search at Icheon logistics warehouse fire site.\"",
"The image depicts individuals in white protective suits and helmets, equipped with gas masks, walking through a debris-filled area. The visible text in Korean at the bottom of the image translates to \"Police, detailed search at Icheon logistics warehouse fire site.\" This suggests that these individuals are conducting a thorough investigation at the site of a fire at a logistics warehouse in Icheon.",
"The current frame shows a large excavator working near a charred building, likely part of the logistics warehouse complex. The visible text in Korean at the bottom translates to \"Jeonghal, Icheon fire and explosion, detailed search for the deceased.\" This indicates that the excavator is assisting in the search efforts for any remaining individuals who may have been affected by the fire and explosion at the Icheon site.",
"The scene continues to focus on the large excavator actively engaged in demolition or search efforts near the severely damaged building. The excavator's claw is extended towards the remains, indicating ongoing work to clear debris or locate any individuals affected by the incident. The text \"정할, 이천 물류창고 화재현장 정밀 수색\" (Jeonghal, Icheon Logistics Warehouse Fire Scene Detailed Search) remains visible, confirming the continuity of the search operation."
] |
n8dcNr14bm8
|
A group of individuals, including police, in protective suits and masks, are conducting a detailed search at the site of a fire and explosion at a logistics warehouse in Icheon. The search efforts involve a large excavator that is assisting in clearing debris and locating any remaining individuals who may have been affected by the incident. The team is working through a debris-filled area, navigating rubble and charred remains of the warehouse, in an attempt to thoroughly investigate the site and account for anyone involved in the accident.
|
A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.
|
[
"Flames engulfing the construction site in Icheon, South Korea on April 29, 2020, during a fire incident",
"Firefighters battling the blaze with water hoses during the Icheon, South Korea fire on 29 April 2020",
"Emergency responders rushing to rescue trapped workers during the Icheon, South Korea fire on April 29, 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Aerial footage of the construction site fire in Icheon, South Korea showing the extent of the damage on 29 April 2020"
] |
[
"Firefighters battling the blaze at the construction site in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing people trapped in the burning building during the Icheon, South Korea fire on 29 April 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Police investigating the cause of the fire in Icheon, South Korea on 29 April 2020",
"Rescue teams searching for missing people in the rubble after the Icheon, South Korea fire on 29 April 2020"
] |
[
"Flames engulfing the construction site in Icheon, South Korea during the fire on 29 April 2020",
"People running away from the burning building during the Icheon fire in South Korea on April 29, 2020",
"Firefighters arriving at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing injured individuals from a fire in Icheon, South Korea on 29 April 2020",
"Smoke billowing into the air as the fire rages on in Icheon, South Korea on 29 April 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%EC%9D%B4%EC%B2%9C_%EB%AC%BC%EB%A5%98%EC%84%BC%ED%84%B0_%EA%B3%B5%EC%82%AC%EC%9E%A5_%ED%99%94%EC%9E%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, as these would be the most visual and relevant aspects of the incident. The extracted events should focus on the fire, rescue efforts, and potential damage, as these are the most likely to be captured on video.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. People running away from the burning building\n3. Firefighters arriving at the scene\n4. Emergency responders rescuing injured individuals\n5. Smoke billowing into the air as the fire rages on",
"en_article_excerpt": "A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Icheon_fire",
"event_category": "disasters",
"event_name": "icheon_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the typical scenes that could be visualized in a video associated with a major fire at a construction site, such as the fire itself, rescue efforts, and the aftermath. Given the location and context of the event, it is likely that the extracted events will involve emergency responders, construction equipment, and damaged buildings.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. Firefighters battling the blaze with water hoses\n3. Emergency responders rushing to rescue trapped workers\n4. Paramedics treating injured people at the scene\n5. Aerial footage of the construction site showing the extent of the damage",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage assessment, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the construction site\n2. Emergency responders rescuing people trapped in the burning building\n3. Paramedics treating injured people at the scene\n4. Police investigating the cause of the fire\n5. Rescue teams searching for missing people in the rubble",
"video_URL": "http://youtube.com/watch?v=Z1TmlndGWrg",
"video_description": "대규모 인명 피해가 발생한 경기 이천 물류센터 공사장 화재와 관련해 경찰이 전담 수사본부를 편성했습니다. 경기 남부지방경찰청은 광역수사대와 과학수사대 등 125명 규모의 전담 수사본부를 편성했다고 밝혔습니다. 수사본부는 오늘(30일) 오전 10시 반 국립과학수사연구원, 소방당국과 함께 합동으로 현장감식을 진행할 예정입니다. 또, 공사 과정에서 소방법이나 건축법을 위반한 사항이 있는지 확인할 방침입니다. 신준명[shinjm7529@ytn.co.kr] ※ '당신의 제보가 뉴스가 됩니다' YTN은 여러분의 소중한 제보를 기다립니다. [카카오톡] YTN을 검색해 채널 추가 [전화] 02-398-8585 [메일] social@ytn.co.kr [온라인 제보] www.ytn.co.kr ▶ 기사 원문 : https://www.ytn.co.kr/_ln/0103_202004300027155233 ▶ 제보 안내 : http://goo.gl/gEvsAL, 모바일앱, social@ytn.co.kr, #2424 ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN & YTN PLUS 무단 전재 및 재배포금지",
"video_id": "Z1TmlndGWrg"
}
| 4
|
[
"Firefighters and emergency personnel are gathered around a fire truck, responding to an incident at a building entrance that appears to have been involved in a fire or explosion. There is visible damage and debris near the entrance.",
"The current frame shows the aftermath of a fire or explosion at a building, with significant damage evident on the exterior. A large section of a window is missing, leaving jagged edges and a darkened, charred area around the opening. The text on the image reads '이천 물류센터 화재 수사본부 폐쇄… 오늘 합동감식' (Translation: 'Icheon Logistics Center fire investigation headquarters closed… Joint inspection today'), indicating that the incident is being investigated and a joint inspection is scheduled for the day. The focus on the damaged window highlights the extent of the damage caused by the incident.",
"The current frame continues to show the aftermath of a fire or explosion at the same building, with the focus now on a different section of the exterior. The building's surface is heavily damaged, with large sections of material missing and charred areas evident. The text on the image reads '이천 물류센터 화재 수사본부 폐쇄… 오늘 합동감식' (Translation: 'Icheon Logistics Center fire investigation headquarters closed… Joint inspection today'), which remains consistent with the previous frame and indicates ongoing investigation activities. The continuity in the text suggests that the video is providing updates on the investigation of the incident, emphasizing the severity of the damage caused.",
"The scene depicts a group of responders, including emergency services and possibly investigators, gathered at the site of an incident involving a collapsed building. The individuals are dressed in protective gear and some are wearing helmets, indicating a hazardous environment. Several ambulances are present, suggesting the possibility of injuries or ongoing medical response. The responders are focused on the area of the collapse, and there is a sense of urgency and coordination in their actions. The presence of multiple responders and emergency vehicles points to a significant event requiring immediate attention."
] |
Z1TmlndGWrg
|
A fire or explosion occurred at the Icheon Logistics Center, causing significant damage to the building's exterior, including broken windows and charred areas. Emergency responders and investigators are on the scene, with firefighters, ambulances, and other emergency services present. The incident is under investigation, with a joint inspection scheduled for the day, and the investigation headquarters has been closed. The responders are working together to assess the damage and respond to the incident, with a sense of urgency and coordination in their actions, suggesting a potentially hazardous environment and possible injuries.
|
A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.
|
[
"Flames engulfing the construction site in Icheon, South Korea on April 29, 2020, during a fire incident",
"Firefighters battling the blaze with water hoses during the Icheon, South Korea fire on 29 April 2020",
"Emergency responders rushing to rescue trapped workers during the Icheon, South Korea fire on April 29, 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Aerial footage of the construction site fire in Icheon, South Korea showing the extent of the damage on 29 April 2020"
] |
[
"Firefighters battling the blaze at the construction site in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing people trapped in the burning building during the Icheon, South Korea fire on 29 April 2020",
"Paramedics treating injured people at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Police investigating the cause of the fire in Icheon, South Korea on 29 April 2020",
"Rescue teams searching for missing people in the rubble after the Icheon, South Korea fire on 29 April 2020"
] |
[
"Flames engulfing the construction site in Icheon, South Korea during the fire on 29 April 2020",
"People running away from the burning building during the Icheon fire in South Korea on April 29, 2020",
"Firefighters arriving at the scene of a fire in Icheon, South Korea on April 29, 2020",
"Emergency responders rescuing injured individuals from a fire in Icheon, South Korea on 29 April 2020",
"Smoke billowing into the air as the fire rages on in Icheon, South Korea on 29 April 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%EC%9D%B4%EC%B2%9C_%EB%AC%BC%EB%A5%98%EC%84%BC%ED%84%B0_%EA%B3%B5%EC%82%AC%EC%9E%A5_%ED%99%94%EC%9E%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, as these would be the most visual and relevant aspects of the incident. The extracted events should focus on the fire, rescue efforts, and potential damage, as these are the most likely to be captured on video.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. People running away from the burning building\n3. Firefighters arriving at the scene\n4. Emergency responders rescuing injured individuals\n5. Smoke billowing into the air as the fire rages on",
"en_article_excerpt": "A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Icheon_fire",
"event_category": "disasters",
"event_name": "icheon_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the typical scenes that could be visualized in a video associated with a major fire at a construction site, such as the fire itself, rescue efforts, and the aftermath. Given the location and context of the event, it is likely that the extracted events will involve emergency responders, construction equipment, and damaged buildings.\n\nEVENTS:\n1. Flames engulfing the construction site\n2. Firefighters battling the blaze with water hoses\n3. Emergency responders rushing to rescue trapped workers\n4. Paramedics treating injured people at the scene\n5. Aerial footage of the construction site showing the extent of the damage",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: A major fire occurred on 29 April 2020 at a construction site in Icheon, South Korea, killing 39 people and injuring another 10.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a major fire at a construction site in Icheon, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage assessment, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the construction site\n2. Emergency responders rescuing people trapped in the burning building\n3. Paramedics treating injured people at the scene\n4. Police investigating the cause of the fire\n5. Rescue teams searching for missing people in the rubble",
"video_URL": "http://youtube.com/watch?v=tHSh4jwdIJg",
"video_description": "29일 오후 1시 반쯤 경기도 이천시 모가면의 한 물류창고 공사현장에서 불이 났습니다. 이 불로 오후 5시 기준 25명이 숨진 것으로 소방당국은 추정하고 있습니다. 불이 났을 당시 현장에서 9개 업체, 80여 명이 작업 중이던 것으로 알려져 인명 피해가 더 커질 가능성도 있는 상황입니다. 물류창고 지하 2층에서 불이 시작된 것으로 추정하며, 소방당국은 \"불이 빠르게 확산되는 것으로 봐서 가연성 물질인 우레탄폼 작업 중 화재가 난 것으로 추정한다\"고 말했습니다. 현장 영상을 전해드립니다. 이천 물류창고 화재 현장 상황…소방당국 \"우레탄폼 원인 추정\" / 비디오머그 ✔신뢰할 수 있는 재미, 비디오머그 VIDEOMUG #비디오머그 #이천 #물류창고_화재 ▶비디오머그 채널 구독하기 : https://n.sbs.co.kr/mugTube ▶Homepage : https://n.sbs.co.kr/videomug ▶Facebook : https://www.facebook.com/videomug ▶Twitter : https://twitter.com/videomug_sbs",
"video_id": "tHSh4jwdIJg"
}
| 4
|
[
"The frame shows a large building engulfed in thick black smoke, indicating a fire. Several people, possibly construction workers or rescue personnel, are gathered outside the building, some holding hoses or other equipment, likely attempting to control the fire. There is a white vehicle parked nearby, and construction materials and equipment are visible on the ground. The text on the image indicates a fire at Soeongbundanggu, with 25 people trapped.",
"The current frame shows the same large building still engulfed in thick black smoke, indicating an ongoing fire. Several people, likely construction workers or rescue personnel, are still gathered outside the building, some holding hoses or other equipment, continuing their efforts to control the fire. A white vehicle is still parked nearby, and construction materials and equipment are visible on the ground. The text on the image confirms the fire at Soeongbundanggu with 25 people trapped.",
"The current frame shows a large building still engulfed in thick black smoke, indicating an ongoing fire. A significant amount of smoke is billowing into the sky. In the foreground, there is a sports field with a track and a parking lot filled with cars. The text on the image confirms the fire at Soeongbundanggu, stating that the fire is almost completely extinguished but that it is difficult for rescue vehicles to approach the site due to the smoke.",
"The current frame continues to depict the large building engulfed in thick black smoke, indicating the fire is still ongoing. The smoke billows heavily into the sky, obscuring parts of the scene. The sports field with a track and the parking lot filled with cars remains in the foreground. The text on the image reiterates that the fire at Soeongbundanggu is almost completely extinguished, but rescue vehicles are still facing difficulty approaching the site because of the smoke."
] |
tHSh4jwdIJg
|
A large building in Soeongbundanggu is engulfed in a severe fire, with thick black smoke billowing into the sky. Construction workers or rescue personnel are attempting to control the fire, using hoses and equipment, while 25 people are trapped inside. The fire rages on for some time, with rescue efforts hindered by the dense smoke, which also causes difficulties for rescue vehicles to approach the site. Eventually, the fire is almost completely extinguished, but the smoke persists, affecting the surrounding area, including a nearby sports field and parking lot.
|
On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.
|
[
"A person setting fire to the office building in Daegu, South Korea on 9 June 2022",
"A person setting fire to the office building during the arson attack in Daegu, South Korea on 9 June 2022",
"Flames engulfing a building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing the building during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating the office building in a state of panic due to a fire in Daegu, South Korea on 9 June 2022",
"People evacuating the office building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Firefighters arriving at the scene of a fire in Daegu, South Korea on 9 June 2022 to extinguish the blaze",
"Firefighters arriving at the scene to extinguish the fire during the arson attack in Daegu, South Korea on June 9, 2022",
"Paramedics treating injured people outside the building after fire in Daegu, South Korea on 9 June 2022",
"Paramedics treating injured people outside the building after arson attack in Daegu, South Korea on June 9, 2022"
] |
[
"Firefighters battling the blaze at the office building in Daegu, South Korea on June 9, 2022",
"Firefighters battling the blaze at the office building in Daegu, South Korea after alleged arson attack on 9 June 2022",
"People evacuating a building in a state of panic due to a fire in Daegu, South Korea on June 9, 2022",
"People evacuating the building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Emergency responders providing medical aid to injured victims of the June 9, 2022 fire in Daegu, South Korea",
"Emergency responders providing medical aid to injured victims of the June 9 2022 arson attack in Daegu, South Korea",
"Flames engulfing parts of the office building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing parts of the office building in Daegu, South Korea during the arson attack on 9 June 2022",
"Rescue teams searching for survivors in the building damaged by the fire in Daegu, South Korea on 9 June 2022",
"Rescue teams searching for survivors in the damaged building after the arson attack in Daegu, South Korea on 9 June 2022"
] |
[
"Flames engulfing the office building in Daegu, South Korea on 9 June 2022 fire incident",
"Flames engulfing the office building in Daegu, South Korea due to arson attack on 9 June 2022",
"Firefighters arriving at the scene and battling the blaze in Daegu, South Korea on 9 June 2022 during a large fire incident",
"Firefighters arriving at the scene and battling the blaze during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating building after fire in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"People evacuating building after arson attack in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"Emergency medical personnel treating injured individuals outside the building after fire in Daegu, South Korea on 9 June 2022",
"Emergency medical personnel treating injured individuals outside the building after the arson attack in Daegu, South Korea on June 9, 2022",
"Damage to the office building in Daegu, South Korea after the fire on 9 June 2022 has been extinguished",
"Damage to the office building in Daegu, South Korea after the arson attack fire was extinguished on 9 June 2022"
] |
{
"article_url": "https://namu.wiki/w/%EB%8C%80%EA%B5%AC%20%EB%B3%80%ED%98%B8%EC%82%AC%20%EC%82%AC%EB%AC%B4%EC%8B%A4%20%EB%B0%A9%ED%99%94%20%EC%82%AC%EA%B1%B4",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: To extract events from the given query, it makes sense to focus on the concrete and visual aspects of the fire and its aftermath, considering what could be reasonably captured in a video. This includes the fire itself, the response of emergency services, and the impact on the building and its occupants.\n\nEVENTS:\n1. Flames engulfing the office building in Daegu\n2. Firefighters arriving at the scene and battling the blaze\n3. People evacuating the building, some injured and covered in soot\n4. Emergency medical personnel treating injured individuals outside the building\n5. Damage to the office building after the fire has been extinguished",
"en_article_excerpt": "On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.",
"en_article_url": "https://en.wikipedia.org/wiki/Daegu_office_fire",
"event_category": "disasters",
"event_name": "daegu_office_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the actions and consequences leading up to and resulting from the arson attack, such as the fire breaking out, people evacuating, and emergency responders arriving. These events are concrete, visual, and likely to be captured in a video related to the incident.\n\nEVENTS:\n1. A person setting fire to the office building\n2. Flames engulfing the building\n3. People evacuating the office building in a state of panic\n4. Firefighters arriving at the scene to extinguish the fire\n5. Paramedics treating injured people outside the building",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a fire at an office building in Daegu, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage to the building, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the office building\n2. People evacuating the building in a state of panic\n3. Emergency responders providing medical aid to injured victims\n4. Flames engulfing parts of the office building\n5. Rescue teams searching for survivors in the damaged building",
"video_URL": "http://youtube.com/watch?v=I2CxZCzE5pE",
"video_description": "지금부턴 어제(9일) 대구의 한 변호사 사무실에서 난 화재 관련해서 저희가 새롭게 취재한 내용 전해드리겠습니다. 저희 취재진이 사고가 난 건물의 내부 영상을 입수했습니다. 불이 시작된 건물 2층은 형체를 알아보기 어려울 정도로 검게 타버린 모습이었습니다. ▶ 기사 전문 https://news.jtbc.joins.com/html/151/NB12062151.html ▶ 뉴스룸 다시보기 (https://bit.ly/2nxI8jQ) #이승환기자 #JTBC뉴스룸 ☞JTBC뉴스레터 구독하기 (https://bit.ly/3IOFY5H) ☞JTBC유튜브 구독하기 (https://bit.ly/2hYgWZg) ☞JTBC유튜브 커뮤니티 (https://bit.ly/2LZIwke) #JTBC뉴스 공식 페이지 (홈페이지) https://news.jtbc.joins.com (APP) https://bit.ly/1r04W2D 페이스북 https://www.facebook.com/jtbcnews 트위터 https://twitter.com/JTBC_news 인스타그램 https://www.instagram.com/jtbcnews ☏ 제보하기 https://bit.ly/1krluzF 방송사 : JTBC (https://jtbc.joins.com)",
"video_id": "I2CxZCzE5pE"
}
| 4
|
[
"The image shows a close-up of several steps that appear to be damaged, with blackened and cracked surfaces, indicating possible water damage or wear. The text in the top left corner is in Korean and translates to \"Recently discovered 'gum floor'\". The logo of JTBC is visible in the top right corner.",
"The image shows an empty, dimly lit indoor area with a white pillar on the left and a glass-enclosed structure on the right. The text in the top left corner is in Korean and translates to \"Recently discovered 'gum floor'\". The logo of JTBC is visible in the top right corner. The text at the bottom of the image reads, \"The people who disappeared mysteriously, now return and are found.\" This suggests a narrative shift from the previous frame, which focused on the physical condition of the steps, to the mysterious reappearance of people.",
"The image depicts a group of people outdoors, gathered around a reporter who is holding a microphone and speaking. The setting appears to be an urban street with buildings and trees in the background. The text at the top of the image is in Korean and translates to \"Recently discovered 'gum floor'\". The text at the bottom of the image reads, \"We're experiencing a sudden drop in temperature, and just a moment ago, a 20~30-minute rainstorm. There were about 5 people who were outside at the time.\" This suggests that the group is discussing or reacting to recent weather conditions and possibly the impact on their activities or the environment.",
"The image shows a news studio with a large screen displaying a city skyline at dusk. The JTBC News logo is prominently displayed in the foreground. Korean text on the screen translates to \"During this show, we will provide more news.\" The setting appears to be a professional news broadcast studio, indicating a shift from an outdoor scene to an indoor news reporting environment, likely discussing recent events or news updates."
] |
I2CxZCzE5pE
|
A news report by JTBC covers the mysterious reappearance of people who had disappeared, with the story beginning with an introduction to a "gum floor" that shows signs of water damage or wear. The scene shifts to an empty indoor area, then to an outdoor setting where a reporter is discussing recent weather conditions, including a sudden drop in temperature and a brief rainstorm, with a group of people gathered around. The report then transitions to a news studio, where further news and updates on the events are promised to be provided, suggesting an ongoing investigation or coverage of the mysterious disappearances and reappearances, potentially linked to the unusual weather conditions and the condition of the "gum floor".
|
On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.
|
[
"A person setting fire to the office building in Daegu, South Korea on 9 June 2022",
"A person setting fire to the office building during the arson attack in Daegu, South Korea on 9 June 2022",
"Flames engulfing a building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing the building during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating the office building in a state of panic due to a fire in Daegu, South Korea on 9 June 2022",
"People evacuating the office building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Firefighters arriving at the scene of a fire in Daegu, South Korea on 9 June 2022 to extinguish the blaze",
"Firefighters arriving at the scene to extinguish the fire during the arson attack in Daegu, South Korea on June 9, 2022",
"Paramedics treating injured people outside the building after fire in Daegu, South Korea on 9 June 2022",
"Paramedics treating injured people outside the building after arson attack in Daegu, South Korea on June 9, 2022"
] |
[
"Firefighters battling the blaze at the office building in Daegu, South Korea on June 9, 2022",
"Firefighters battling the blaze at the office building in Daegu, South Korea after alleged arson attack on 9 June 2022",
"People evacuating a building in a state of panic due to a fire in Daegu, South Korea on June 9, 2022",
"People evacuating the building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Emergency responders providing medical aid to injured victims of the June 9, 2022 fire in Daegu, South Korea",
"Emergency responders providing medical aid to injured victims of the June 9 2022 arson attack in Daegu, South Korea",
"Flames engulfing parts of the office building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing parts of the office building in Daegu, South Korea during the arson attack on 9 June 2022",
"Rescue teams searching for survivors in the building damaged by the fire in Daegu, South Korea on 9 June 2022",
"Rescue teams searching for survivors in the damaged building after the arson attack in Daegu, South Korea on 9 June 2022"
] |
[
"Flames engulfing the office building in Daegu, South Korea on 9 June 2022 fire incident",
"Flames engulfing the office building in Daegu, South Korea due to arson attack on 9 June 2022",
"Firefighters arriving at the scene and battling the blaze in Daegu, South Korea on 9 June 2022 during a large fire incident",
"Firefighters arriving at the scene and battling the blaze during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating building after fire in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"People evacuating building after arson attack in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"Emergency medical personnel treating injured individuals outside the building after fire in Daegu, South Korea on 9 June 2022",
"Emergency medical personnel treating injured individuals outside the building after the arson attack in Daegu, South Korea on June 9, 2022",
"Damage to the office building in Daegu, South Korea after the fire on 9 June 2022 has been extinguished",
"Damage to the office building in Daegu, South Korea after the arson attack fire was extinguished on 9 June 2022"
] |
{
"article_url": "https://namu.wiki/w/%EB%8C%80%EA%B5%AC%20%EB%B3%80%ED%98%B8%EC%82%AC%20%EC%82%AC%EB%AC%B4%EC%8B%A4%20%EB%B0%A9%ED%99%94%20%EC%82%AC%EA%B1%B4",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: To extract events from the given query, it makes sense to focus on the concrete and visual aspects of the fire and its aftermath, considering what could be reasonably captured in a video. This includes the fire itself, the response of emergency services, and the impact on the building and its occupants.\n\nEVENTS:\n1. Flames engulfing the office building in Daegu\n2. Firefighters arriving at the scene and battling the blaze\n3. People evacuating the building, some injured and covered in soot\n4. Emergency medical personnel treating injured individuals outside the building\n5. Damage to the office building after the fire has been extinguished",
"en_article_excerpt": "On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.",
"en_article_url": "https://en.wikipedia.org/wiki/Daegu_office_fire",
"event_category": "disasters",
"event_name": "daegu_office_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the actions and consequences leading up to and resulting from the arson attack, such as the fire breaking out, people evacuating, and emergency responders arriving. These events are concrete, visual, and likely to be captured in a video related to the incident.\n\nEVENTS:\n1. A person setting fire to the office building\n2. Flames engulfing the building\n3. People evacuating the office building in a state of panic\n4. Firefighters arriving at the scene to extinguish the fire\n5. Paramedics treating injured people outside the building",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a fire at an office building in Daegu, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage to the building, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the office building\n2. People evacuating the building in a state of panic\n3. Emergency responders providing medical aid to injured victims\n4. Flames engulfing parts of the office building\n5. Rescue teams searching for survivors in the damaged building",
"video_URL": "http://youtube.com/watch?v=oVzTD-Qy5HM",
"video_description": "대구 변호사 사무실 빌딩 방화 추정 화재…7명 사망 [앵커] 오늘(9일) 낮 대구 지법 근처의 한 변호사 사무실에서 방화로 보이는 화재가 발생해, 방화 용의자 1명을 포함해 7명이 숨지고 40여 명이 다쳤습니다. 현장 취재기자 연결해 자세한 소식 들어봅니다. 정지훈 기자. [기자] 네, 대구 변호사 사무실 빌딩 화재 현장에 나와 있습니다. 제 바로 뒤편 건물을 보시면 전면 유리창들이 깨진 걸 보실 수 있을텐데요. 신고를 받고 출동한 소방당국이 화재 진압과 구조를 위해 일부러 깨드린 겁니다. 당시 상황이 얼마나 긴박했는지를 알 수 있습니다. 이 건물은 지하 2층, 지상 5층짜리 건물로 오늘(9일) 오전 10시 55분쯤 건물 2층에서 불이 났습니다. 불은 소방 당국에 의해 20여 분만에 꺼졌지만, 남자 5명, 여자 2명 등 7명이 숨지고 40여 명이 화상 또는 연기를 흡입해 병원 치료를 받고 있습니다. 이들 사망자는 2층 203호에서 모두 발견됐는데요. 화재 발생 당시 119에는 \"건물 2층에서 검은 연기가 보이고 폭발음이 들렸다\"는 신고가 접수됐습니다. 사망자의 신원이 아직 확인되지 않았습니다. 같은 장소에서 방화 용의자도 숨진 채 발견됐습니다. 불이 나자 건물 3층에 있던 사람들은 4층 등 건물 위층으로, 일부는 건물 밖에 설치된 비상계단을 통해 옥상으로 대피하기도 했습니다. 화재 현장에 있었던 목격자의 말 들어보시겠습니다. [현장 목격자] \"어느 순간 폭발음이 팡 터지면서 유리창 깨지는 소리가 계속 나고 제가 너무 놀라서 상담실에서 뛰쳐 나와서 문을 열려고 하니 이미 뜨거웠어요. 그래서 몸으로 문을 밀쳐서 열고…\" 소방당국은 화재 진압 과정에서 급격히 연소가 확대돼 인명 피해 규모가 커졌다고 설명했습니다. 당국은 브리핑에서 \"화재 건물 지하층에만 스프링클러가 있고 지상층에는 스프링클러가 설치돼 있지 않다\"고 밝혔습니다. 화재가 발생한 빌딩은 법원 뒤쪽에 위치해 변호사 사무실이 밀집한 곳입니다. 경찰은 50대 방화 용의자가 CCTV상 주거지에서 뭔가를 들고 나오는 장면을 확인하고 자세한 화재 경위를 조사하고 있는데요. 경찰은 불이 난 변호사 사무실에 의뢰된 송사 사건의 상대인 용의자가 사무실에 불을 낸 것으로 보고, 수사전담팀을 편성해 사건 발생 경위 등을 집중 조사할 방침입니다. 지금까지 대구에서 연합뉴스TV 정지훈입니다. #대구_변호사_사무실 #폭발 #방화 연합뉴스TV 기사문의 및 제보 : 카톡/라인 jebo23 ▣ 연합뉴스TV 유튜브 채널 구독 https://goo.gl/VuCJMi ▣ 대한민국 뉴스의 시작 연합뉴스TV / Yonhap News TV http://www.yonhapnewstv.co.kr/",
"video_id": "oVzTD-Qy5HM"
}
| 4
|
[
"A reporter from 연합뉴스 TV is standing outside near a damaged building, speaking into a microphone. The scene is chaotic with people, some in uniforms, moving around. The text on the screen indicates a building explosion, with seven casualties reported, and mentions a court hearing on the ninth at 9:49.",
"The current frame shows a group of uniformed individuals, possibly police or emergency personnel, standing and moving in an area. The focus is on a person in uniform talking on a phone, suggesting coordination or communication is taking place. The background includes more uniformed personnel and a partial view of emergency vehicles, reinforcing the sense of an active response to the situation. The visible text states that the incident involves the participation of over 400 million won and 400 million won in criminal damage, indicating the financial and legal aspects of the incident. The time displayed is 16:49.",
"The current frame shows a wide shot of a building that has sustained visible damage, with debris and broken windows indicating a recent incident. The setting appears to be an urban environment with trees and other buildings in the background. The visible text suggests that seven people are missing and that the event is related to a building collapse, specifically involving a collapsed residential area in a city. The time displayed is 16:50.",
"The current frame shows a man standing outside a building that has sustained damage, indicated by broken windows and debris. He is wearing a white shirt, a mask, and holding a microphone with the text \"연합뉴스\" (Yonhap News). There are several people around, including uniformed officers and bystanders observing the situation. The visible text \"16:50\" and \"신뢰 보여주는 건...\" (Showing Trust...) suggests the ongoing coverage of the incident. The scene continues to depict the aftermath of the building collapse in an urban setting."
] |
oVzTD-Qy5HM
|
A building explosion and collapse occurred in an urban area, resulting in seven reported casualties and missing persons. Emergency responders, including over 400 personnel, are on the scene, with a significant financial loss of over 400 million won in damages. A court hearing is scheduled for the ninth. Reporter coverage from 연합뉴스 TV shows the chaotic scene, with debris, broken windows, and uniformed individuals coordinating the response. The incident is under investigation, with the focus on the collapsed residential area and the ongoing efforts to address the situation.
|
On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.
|
[
"A person setting fire to the office building in Daegu, South Korea on 9 June 2022",
"A person setting fire to the office building during the arson attack in Daegu, South Korea on 9 June 2022",
"Flames engulfing a building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing the building during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating the office building in a state of panic due to a fire in Daegu, South Korea on 9 June 2022",
"People evacuating the office building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Firefighters arriving at the scene of a fire in Daegu, South Korea on 9 June 2022 to extinguish the blaze",
"Firefighters arriving at the scene to extinguish the fire during the arson attack in Daegu, South Korea on June 9, 2022",
"Paramedics treating injured people outside the building after fire in Daegu, South Korea on 9 June 2022",
"Paramedics treating injured people outside the building after arson attack in Daegu, South Korea on June 9, 2022"
] |
[
"Firefighters battling the blaze at the office building in Daegu, South Korea on June 9, 2022",
"Firefighters battling the blaze at the office building in Daegu, South Korea after alleged arson attack on 9 June 2022",
"People evacuating a building in a state of panic due to a fire in Daegu, South Korea on June 9, 2022",
"People evacuating the building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Emergency responders providing medical aid to injured victims of the June 9, 2022 fire in Daegu, South Korea",
"Emergency responders providing medical aid to injured victims of the June 9 2022 arson attack in Daegu, South Korea",
"Flames engulfing parts of the office building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing parts of the office building in Daegu, South Korea during the arson attack on 9 June 2022",
"Rescue teams searching for survivors in the building damaged by the fire in Daegu, South Korea on 9 June 2022",
"Rescue teams searching for survivors in the damaged building after the arson attack in Daegu, South Korea on 9 June 2022"
] |
[
"Flames engulfing the office building in Daegu, South Korea on 9 June 2022 fire incident",
"Flames engulfing the office building in Daegu, South Korea due to arson attack on 9 June 2022",
"Firefighters arriving at the scene and battling the blaze in Daegu, South Korea on 9 June 2022 during a large fire incident",
"Firefighters arriving at the scene and battling the blaze during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating building after fire in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"People evacuating building after arson attack in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"Emergency medical personnel treating injured individuals outside the building after fire in Daegu, South Korea on 9 June 2022",
"Emergency medical personnel treating injured individuals outside the building after the arson attack in Daegu, South Korea on June 9, 2022",
"Damage to the office building in Daegu, South Korea after the fire on 9 June 2022 has been extinguished",
"Damage to the office building in Daegu, South Korea after the arson attack fire was extinguished on 9 June 2022"
] |
{
"article_url": "https://namu.wiki/w/%EB%8C%80%EA%B5%AC%20%EB%B3%80%ED%98%B8%EC%82%AC%20%EC%82%AC%EB%AC%B4%EC%8B%A4%20%EB%B0%A9%ED%99%94%20%EC%82%AC%EA%B1%B4",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: To extract events from the given query, it makes sense to focus on the concrete and visual aspects of the fire and its aftermath, considering what could be reasonably captured in a video. This includes the fire itself, the response of emergency services, and the impact on the building and its occupants.\n\nEVENTS:\n1. Flames engulfing the office building in Daegu\n2. Firefighters arriving at the scene and battling the blaze\n3. People evacuating the building, some injured and covered in soot\n4. Emergency medical personnel treating injured individuals outside the building\n5. Damage to the office building after the fire has been extinguished",
"en_article_excerpt": "On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.",
"en_article_url": "https://en.wikipedia.org/wiki/Daegu_office_fire",
"event_category": "disasters",
"event_name": "daegu_office_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the actions and consequences leading up to and resulting from the arson attack, such as the fire breaking out, people evacuating, and emergency responders arriving. These events are concrete, visual, and likely to be captured in a video related to the incident.\n\nEVENTS:\n1. A person setting fire to the office building\n2. Flames engulfing the building\n3. People evacuating the office building in a state of panic\n4. Firefighters arriving at the scene to extinguish the fire\n5. Paramedics treating injured people outside the building",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a fire at an office building in Daegu, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage to the building, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the office building\n2. People evacuating the building in a state of panic\n3. Emergency responders providing medical aid to injured victims\n4. Flames engulfing parts of the office building\n5. Rescue teams searching for survivors in the damaged building",
"video_URL": "http://twitter.com/i/status/1534839444830044161",
"video_description": "대구 변호사 사무실 화재",
"video_id": "1534839444830044161"
}
| 4
|
[
"The image shows a building surrounded by trees, with a large cloud of smoke rising from its roof. There is a person visible on the roof, seemingly inspecting or observing the situation. The text in the image indicates that this is a broadcast by KBS1, and it mentions a timestamp of \"오전 10시 55분\" (10:55 AM) and a possible fire or smoke incident. The context suggests an emergency or inspection situation, possibly related to a building malfunction or accident.",
"The current frame depicts a group of individuals in high-visibility clothing, possibly firefighters or rescue workers, on what appears to be a bridge or elevated walkway. They are gathered around a ladder, which is leaning against a structure, indicating that they are involved in an emergency response or inspection. The presence of the ladder suggests that they might be assessing or accessing the roof area of a building, which is consistent with the previous frame where a person was seen on the roof. The text on the screen, including \"KBS1\" and a reference to \"불락리스트 의혹 백운규 전 산업부 장관 소환\" (Indictment of former Minister of Industry Bae Kun-gyu on blacklist suspicion), implies that this is part of a broadcast news segment. The situation appears to be part of an ongoing emergency or investigative response to an incident involving the building shown in the previous frame.",
"The current frame shows significant damage to a building's window, with large cracks and shattered glass, indicating that an explosion or impact has occurred. The text on the screen, \"판결 불만 방화 추정\" (Arson suspected due to dissatisfaction with the verdict), suggests that this damage is related to the aftermath of an event or decision that led to arson. The building's facade reflects the sky, and a ladder is visible, possibly used by the emergency responders seen in the previous frame. This frame is part of a broadcast news segment, as indicated by the \"KBS1\" logo and the text at the bottom, which reads \"KBS뉴스 법무부 촉법소년 연령 하향 검토…한동훈 '소년범죄부터 국민 보호'\" (KBS News: Ministry of Justice reviewing the reduction of the age for juvenile offenders…Han Dong-hoon: Protecting citizens from juvenile crimes). The overall scene depicts the aftermath of an incident involving the building, with ongoing investigation or response efforts.",
"The video frames show a news broadcast featuring a male news anchor in a studio and a female reporter in front of a building. The male anchor is presenting information, while the female reporter is outside the \"우정법원법당\" (Friendship Court House). The text on the screen, \"판결 불만 방화 추정\" (Arson suspected due to dissatisfaction with the verdict), suggests that there is a news story about an arson incident related to dissatisfaction with a court decision. The reporter is holding a microphone with the \"KBS\" logo, indicating she is part of the news team providing live updates on the situation. The lower text reads \"타이완 유학생 살인' 음주 운전자, 윤창호법 위헌에도 징역 8년 확정\" (Taiwanese university student's murder, driver under influence, despite the Yoon Chang-ho law being unconstitutional, sentenced to 8 years), which relates to another legal case being covered. The overall scene is a news segment reporting on legal matters and the aftermath of the arson incident."
] |
1534839444830044161
|
A building is involved in an emergency situation, with smoke rising from its roof and a person inspecting the area. Emergency responders, possibly firefighters, are gathered on a bridge or elevated walkway, using a ladder to access the roof. The incident appears to be related to arson, suspected to have been committed due to dissatisfaction with a court verdict, as indicated by significant damage to a window. The event is being covered by a news broadcast on KBS1, with a news anchor in the studio and a reporter providing live updates from outside the Friendship Court House. The news segment also touches on other legal matters, including a case of a driver sentenced to 8 years for murdering a Taiwanese university student while under the influence. Overall, the video depicts an emergency response to an arson incident and its connection to a court case, as well as broader coverage of legal news.
|
On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.
|
[
"A person setting fire to the office building in Daegu, South Korea on 9 June 2022",
"A person setting fire to the office building during the arson attack in Daegu, South Korea on 9 June 2022",
"Flames engulfing a building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing the building during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating the office building in a state of panic due to a fire in Daegu, South Korea on 9 June 2022",
"People evacuating the office building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Firefighters arriving at the scene of a fire in Daegu, South Korea on 9 June 2022 to extinguish the blaze",
"Firefighters arriving at the scene to extinguish the fire during the arson attack in Daegu, South Korea on June 9, 2022",
"Paramedics treating injured people outside the building after fire in Daegu, South Korea on 9 June 2022",
"Paramedics treating injured people outside the building after arson attack in Daegu, South Korea on June 9, 2022"
] |
[
"Firefighters battling the blaze at the office building in Daegu, South Korea on June 9, 2022",
"Firefighters battling the blaze at the office building in Daegu, South Korea after alleged arson attack on 9 June 2022",
"People evacuating a building in a state of panic due to a fire in Daegu, South Korea on June 9, 2022",
"People evacuating the building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Emergency responders providing medical aid to injured victims of the June 9, 2022 fire in Daegu, South Korea",
"Emergency responders providing medical aid to injured victims of the June 9 2022 arson attack in Daegu, South Korea",
"Flames engulfing parts of the office building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing parts of the office building in Daegu, South Korea during the arson attack on 9 June 2022",
"Rescue teams searching for survivors in the building damaged by the fire in Daegu, South Korea on 9 June 2022",
"Rescue teams searching for survivors in the damaged building after the arson attack in Daegu, South Korea on 9 June 2022"
] |
[
"Flames engulfing the office building in Daegu, South Korea on 9 June 2022 fire incident",
"Flames engulfing the office building in Daegu, South Korea due to arson attack on 9 June 2022",
"Firefighters arriving at the scene and battling the blaze in Daegu, South Korea on 9 June 2022 during a large fire incident",
"Firefighters arriving at the scene and battling the blaze during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating building after fire in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"People evacuating building after arson attack in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"Emergency medical personnel treating injured individuals outside the building after fire in Daegu, South Korea on 9 June 2022",
"Emergency medical personnel treating injured individuals outside the building after the arson attack in Daegu, South Korea on June 9, 2022",
"Damage to the office building in Daegu, South Korea after the fire on 9 June 2022 has been extinguished",
"Damage to the office building in Daegu, South Korea after the arson attack fire was extinguished on 9 June 2022"
] |
{
"article_url": "https://namu.wiki/w/%EB%8C%80%EA%B5%AC%20%EB%B3%80%ED%98%B8%EC%82%AC%20%EC%82%AC%EB%AC%B4%EC%8B%A4%20%EB%B0%A9%ED%99%94%20%EC%82%AC%EA%B1%B4",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: To extract events from the given query, it makes sense to focus on the concrete and visual aspects of the fire and its aftermath, considering what could be reasonably captured in a video. This includes the fire itself, the response of emergency services, and the impact on the building and its occupants.\n\nEVENTS:\n1. Flames engulfing the office building in Daegu\n2. Firefighters arriving at the scene and battling the blaze\n3. People evacuating the building, some injured and covered in soot\n4. Emergency medical personnel treating injured individuals outside the building\n5. Damage to the office building after the fire has been extinguished",
"en_article_excerpt": "On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.",
"en_article_url": "https://en.wikipedia.org/wiki/Daegu_office_fire",
"event_category": "disasters",
"event_name": "daegu_office_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the actions and consequences leading up to and resulting from the arson attack, such as the fire breaking out, people evacuating, and emergency responders arriving. These events are concrete, visual, and likely to be captured in a video related to the incident.\n\nEVENTS:\n1. A person setting fire to the office building\n2. Flames engulfing the building\n3. People evacuating the office building in a state of panic\n4. Firefighters arriving at the scene to extinguish the fire\n5. Paramedics treating injured people outside the building",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a fire at an office building in Daegu, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage to the building, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the office building\n2. People evacuating the building in a state of panic\n3. Emergency responders providing medical aid to injured victims\n4. Flames engulfing parts of the office building\n5. Rescue teams searching for survivors in the damaged building",
"video_URL": "http://twitter.com/i/status/1534851387024359424",
"video_description": "흰 천 덮은 물건 든 방화 용의자…CCTV로 본 화재 순간 (출처 : JTBC) http://naver.me/xtqfdWvA 대구 방화 용의자 현장에서 숨져...\"재판 결과에 \\’앙심\\’” / YTN https://youtu.be/Msh294y6bO8 재개발투자 8억 반환소송 패소…2심 앞두고 범행",
"video_id": "1534851387024359424"
}
| 4
|
[
"A person is standing in front of an elevator in a building lobby. There is a staircase to the right with a glass railing, and the setting appears to be an indoor public space.",
"The scene shows a group of individuals, including emergency responders in orange uniforms and a person in a blue blazer and glasses, gathered under a blue tent. They appear to be discussing or assessing a situation. The text on the screen reads \"정화 용의자 현장서 발견\" (Suspect found at the scene of the fire) and provides additional context about the incident being related to CCTV footage.",
"The current frame shows a man in a blue blazer and glasses speaking to a reporter in front of the blue tent. The text on the screen reads \"정화 용의자 현장서 발견\" (Suspect found at the scene of the fire). The scene appears to be a press conference or news report where the man is providing information about the incident. The background shows a group of people, including emergency responders in orange uniforms, gathered under the tent.",
"The frame shows a close-up of a modern building with broken windows, indicating damage to the structure. There are visible cracks and shattered glass, suggesting an event that caused the building to sustain significant damage. The building's reflective glass facade shows the surrounding environment, and there are visible wires and a ladder nearby, possibly for maintenance or investigation purposes. The text 'YTN' is visible, indicating that this footage is from a news broadcast."
] |
1534851387024359424
|
A fire incident occurred in a building, and emergency responders were called to the scene. A suspect was found, and authorities gathered under a blue tent to discuss and assess the situation. A man in a blue blazer, possibly an official or investigator, provided information about the incident to a reporter during a press conference. The building sustained significant damage, with broken windows and shattered glass, and an investigation is underway, with footage of the incident being broadcast on news channels, including YTN.
|
On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.
|
[
"A person setting fire to the office building in Daegu, South Korea on 9 June 2022",
"A person setting fire to the office building during the arson attack in Daegu, South Korea on 9 June 2022",
"Flames engulfing a building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing the building during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating the office building in a state of panic due to a fire in Daegu, South Korea on 9 June 2022",
"People evacuating the office building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Firefighters arriving at the scene of a fire in Daegu, South Korea on 9 June 2022 to extinguish the blaze",
"Firefighters arriving at the scene to extinguish the fire during the arson attack in Daegu, South Korea on June 9, 2022",
"Paramedics treating injured people outside the building after fire in Daegu, South Korea on 9 June 2022",
"Paramedics treating injured people outside the building after arson attack in Daegu, South Korea on June 9, 2022"
] |
[
"Firefighters battling the blaze at the office building in Daegu, South Korea on June 9, 2022",
"Firefighters battling the blaze at the office building in Daegu, South Korea after alleged arson attack on 9 June 2022",
"People evacuating a building in a state of panic due to a fire in Daegu, South Korea on June 9, 2022",
"People evacuating the building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Emergency responders providing medical aid to injured victims of the June 9, 2022 fire in Daegu, South Korea",
"Emergency responders providing medical aid to injured victims of the June 9 2022 arson attack in Daegu, South Korea",
"Flames engulfing parts of the office building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing parts of the office building in Daegu, South Korea during the arson attack on 9 June 2022",
"Rescue teams searching for survivors in the building damaged by the fire in Daegu, South Korea on 9 June 2022",
"Rescue teams searching for survivors in the damaged building after the arson attack in Daegu, South Korea on 9 June 2022"
] |
[
"Flames engulfing the office building in Daegu, South Korea on 9 June 2022 fire incident",
"Flames engulfing the office building in Daegu, South Korea due to arson attack on 9 June 2022",
"Firefighters arriving at the scene and battling the blaze in Daegu, South Korea on 9 June 2022 during a large fire incident",
"Firefighters arriving at the scene and battling the blaze during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating building after fire in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"People evacuating building after arson attack in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"Emergency medical personnel treating injured individuals outside the building after fire in Daegu, South Korea on 9 June 2022",
"Emergency medical personnel treating injured individuals outside the building after the arson attack in Daegu, South Korea on June 9, 2022",
"Damage to the office building in Daegu, South Korea after the fire on 9 June 2022 has been extinguished",
"Damage to the office building in Daegu, South Korea after the arson attack fire was extinguished on 9 June 2022"
] |
{
"article_url": "https://namu.wiki/w/%EB%8C%80%EA%B5%AC%20%EB%B3%80%ED%98%B8%EC%82%AC%20%EC%82%AC%EB%AC%B4%EC%8B%A4%20%EB%B0%A9%ED%99%94%20%EC%82%AC%EA%B1%B4",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: To extract events from the given query, it makes sense to focus on the concrete and visual aspects of the fire and its aftermath, considering what could be reasonably captured in a video. This includes the fire itself, the response of emergency services, and the impact on the building and its occupants.\n\nEVENTS:\n1. Flames engulfing the office building in Daegu\n2. Firefighters arriving at the scene and battling the blaze\n3. People evacuating the building, some injured and covered in soot\n4. Emergency medical personnel treating injured individuals outside the building\n5. Damage to the office building after the fire has been extinguished",
"en_article_excerpt": "On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.",
"en_article_url": "https://en.wikipedia.org/wiki/Daegu_office_fire",
"event_category": "disasters",
"event_name": "daegu_office_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the actions and consequences leading up to and resulting from the arson attack, such as the fire breaking out, people evacuating, and emergency responders arriving. These events are concrete, visual, and likely to be captured in a video related to the incident.\n\nEVENTS:\n1. A person setting fire to the office building\n2. Flames engulfing the building\n3. People evacuating the office building in a state of panic\n4. Firefighters arriving at the scene to extinguish the fire\n5. Paramedics treating injured people outside the building",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a fire at an office building in Daegu, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage to the building, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the office building\n2. People evacuating the building in a state of panic\n3. Emergency responders providing medical aid to injured victims\n4. Flames engulfing parts of the office building\n5. Rescue teams searching for survivors in the damaged building",
"video_URL": "http://youtube.com/watch?v=rzyM4QtZNNU",
"video_description": "오늘(9일) 대구광역시 수성구 범어동 대구지방법원 인근 변호사 사무실 빌딩에서 방화로 추정되는 불이 나 7명이 숨지고 49명이 다쳤습니다. 건물 CCTV를 보면 오전 10시 53분 마스크를 쓰고 빌딩에 들어서는 남성이 확인됩니다. 한 손에는 흰 천으로 덮은 확인되지 않은 물체를 든 상태로, 경찰은 천에 덮인 물체가 인화 물질이었을 것으로 보고 있습니다. 이 남성이 해당 변호사 사무실에 들어간 지 23초 만에 불길이 일어나는 모습이 확인됩니다. ▣ KBS 기사 원문보기 : http://news.kbs.co.kr/news/view.do?ncd=5482165 ▣ 제보 하기 ◇ 카카오톡 : 'KBS제보' 검색 ◇ 전화 : 02-781-1234 ◇ 홈페이지 : https://goo.gl/4bWbkG ◇ 이메일 : kbs1234@kbs.co.kr #대구화재 #방화용의자 #CCTV",
"video_id": "rzyM4QtZNNU"
}
| 4
|
[
"The image shows a security camera view of a person in a green jacket and jeans walking down a hallway near a staircase. The person appears to be carrying something, possibly a bag. The hallway has glass railings and a window, and the scene is from a surveillance camera, indicated by the logo in the upper left corner.",
"The current frame shows a person in dark clothing walking up a staircase with a metal railing. The individual appears to be in motion, ascending the stairs. The scene is still captured from a surveillance camera, indicated by the logo in the upper right corner.",
"The current frame depicts a blurred image with the text \"연기로 가득 찬 복도\" at the bottom, suggesting the scene might be related to an area filled with acting or performances. The logo \"KBS 뉴스\" indicates the footage is from a news broadcast.",
"The current frame shows a logo for \"KBS NEWS\" against a blue and grey geometric background, indicating a transition or introduction segment within the news broadcast. There is no visible action or participants, as the focus is on the branding and design elements."
] |
rzyM4QtZNNU
|
A person in a green jacket and jeans is seen walking down a hallway near a staircase, carrying a bag, as captured by a security camera. Meanwhile, another individual in dark clothing is walking up a staircase. The scene then shifts to a news broadcast, first showing a blurred image with Korean text suggesting a performance or acting context, followed by a KBS NEWS logo against a geometric background, indicating a transition or introduction segment within the news program.
|
On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.
|
[
"A person setting fire to the office building in Daegu, South Korea on 9 June 2022",
"A person setting fire to the office building during the arson attack in Daegu, South Korea on 9 June 2022",
"Flames engulfing a building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing the building during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating the office building in a state of panic due to a fire in Daegu, South Korea on 9 June 2022",
"People evacuating the office building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Firefighters arriving at the scene of a fire in Daegu, South Korea on 9 June 2022 to extinguish the blaze",
"Firefighters arriving at the scene to extinguish the fire during the arson attack in Daegu, South Korea on June 9, 2022",
"Paramedics treating injured people outside the building after fire in Daegu, South Korea on 9 June 2022",
"Paramedics treating injured people outside the building after arson attack in Daegu, South Korea on June 9, 2022"
] |
[
"Firefighters battling the blaze at the office building in Daegu, South Korea on June 9, 2022",
"Firefighters battling the blaze at the office building in Daegu, South Korea after alleged arson attack on 9 June 2022",
"People evacuating a building in a state of panic due to a fire in Daegu, South Korea on June 9, 2022",
"People evacuating the building in a state of panic during the arson attack in Daegu, South Korea on June 9, 2022",
"Emergency responders providing medical aid to injured victims of the June 9, 2022 fire in Daegu, South Korea",
"Emergency responders providing medical aid to injured victims of the June 9 2022 arson attack in Daegu, South Korea",
"Flames engulfing parts of the office building in Daegu, South Korea during the fire on 9 June 2022",
"Flames engulfing parts of the office building in Daegu, South Korea during the arson attack on 9 June 2022",
"Rescue teams searching for survivors in the building damaged by the fire in Daegu, South Korea on 9 June 2022",
"Rescue teams searching for survivors in the damaged building after the arson attack in Daegu, South Korea on 9 June 2022"
] |
[
"Flames engulfing the office building in Daegu, South Korea on 9 June 2022 fire incident",
"Flames engulfing the office building in Daegu, South Korea due to arson attack on 9 June 2022",
"Firefighters arriving at the scene and battling the blaze in Daegu, South Korea on 9 June 2022 during a large fire incident",
"Firefighters arriving at the scene and battling the blaze during the arson attack in Daegu, South Korea on June 9, 2022",
"People evacuating building after fire in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"People evacuating building after arson attack in Daegu, South Korea on June 9, 2022, with some injured and covered in soot",
"Emergency medical personnel treating injured individuals outside the building after fire in Daegu, South Korea on 9 June 2022",
"Emergency medical personnel treating injured individuals outside the building after the arson attack in Daegu, South Korea on June 9, 2022",
"Damage to the office building in Daegu, South Korea after the fire on 9 June 2022 has been extinguished",
"Damage to the office building in Daegu, South Korea after the arson attack fire was extinguished on 9 June 2022"
] |
{
"article_url": "https://namu.wiki/w/%EB%8C%80%EA%B5%AC%20%EB%B3%80%ED%98%B8%EC%82%AC%20%EC%82%AC%EB%AC%B4%EC%8B%A4%20%EB%B0%A9%ED%99%94%20%EC%82%AC%EA%B1%B4",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: To extract events from the given query, it makes sense to focus on the concrete and visual aspects of the fire and its aftermath, considering what could be reasonably captured in a video. This includes the fire itself, the response of emergency services, and the impact on the building and its occupants.\n\nEVENTS:\n1. Flames engulfing the office building in Daegu\n2. Firefighters arriving at the scene and battling the blaze\n3. People evacuating the building, some injured and covered in soot\n4. Emergency medical personnel treating injured individuals outside the building\n5. Damage to the office building after the fire has been extinguished",
"en_article_excerpt": "On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.",
"en_article_url": "https://en.wikipedia.org/wiki/Daegu_office_fire",
"event_category": "disasters",
"event_name": "daegu_office_fire",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the query, it makes sense to consider the actions and consequences leading up to and resulting from the arson attack, such as the fire breaking out, people evacuating, and emergency responders arriving. These events are concrete, visual, and likely to be captured in a video related to the incident.\n\nEVENTS:\n1. A person setting fire to the office building\n2. Flames engulfing the building\n3. People evacuating the office building in a state of panic\n4. Firefighters arriving at the scene to extinguish the fire\n5. Paramedics treating injured people outside the building",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: On 9 June 2022, a fire caused by a suspected arson attack broke out at an office building in Daegu, South Korea, killing seven people and injuring 49 others.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a fire at an office building in Daegu, South Korea, it makes sense to extract events that are directly related to the fire and its aftermath, such as rescue efforts, damage to the building, and emergency responses. These events are likely to be visualized in a video associated with the query, providing a concrete representation of the incident.\n\nEVENTS:\n1. Firefighters battling the blaze at the office building\n2. People evacuating the building in a state of panic\n3. Emergency responders providing medical aid to injured victims\n4. Flames engulfing parts of the office building\n5. Rescue teams searching for survivors in the damaged building",
"video_URL": "http://youtube.com/watch?v=5xiV1Z2MDX4",
"video_description": "대구 변호사 사무실 빌딩 화재…7명 사망·46명 부상 오늘(9일) 오전 11시쯤 대구시 수성구 범어동 대구지방법원 인근 7층짜리 변호사 사무실 빌딩 2층에서 불이 나 7명이 숨지고 46명이 연기를 마시는 등 부상을 입었습니다. 또 안에 있던 수십 명이 긴급 대피했습니다. 사망자는 남성 5명, 여성 2명인 것으로 알려졌습니다. 당시 119에는 \"2층에서 검은 연기가 보이고 폭발음이 들렸다\"는 신고가 접수됐습니다. 불이 나자 소방차 50대와 소방인력 160명이 동원돼 불을 끄고 입주자들을 구조했습니다. 불은 20여 분만에 진화됐습니다. 경찰은 화재 발생 당시 의뢰인이 불만을 제기한 정황이 있었다는 점 등으로 미뤄 방화 여부를 조사하고 있습니다. #사무실_빌딩_화재 #사망 #연기 #대구_범어동 연합뉴스TV 기사문의 및 제보 : 카톡/라인 jebo23 ▣ 연합뉴스TV 유튜브 채널 구독 https://goo.gl/VuCJMi ▣ 대한민국 뉴스의 시작 연합뉴스TV / Yonhap News TV http://www.yonhapnewstv.co.kr/",
"video_id": "5xiV1Z2MDX4"
}
| 4
|
[
"The image shows the exterior of a modern building with reflective glass windows. The text in the image, in Korean, mentions an event involving firefighters in Daejeon. It states that a fire occurred in a bathroom, resulting in the deaths of 7 people and injuries to 46 others. The text is displayed on a news channel, as indicated by the logo in the top right corner. The time displayed is 13:06, and there are additional news tickers at the bottom.",
"The image shows the exterior of a modern building with reflective glass windows. The text in the image, in Korean, mentions an event involving firefighters in Daejeon. It states that a fire occurred in a bathroom, resulting in the deaths of 7 people and injuries to 46 others. The text is displayed on a news channel, as indicated by the logo in the top right corner. The time displayed is 13:06, and there are additional news tickers at the bottom.",
"The current frame continues to show the exterior of the modern building with reflective glass windows. The text in the image, in Korean, reinforces the earlier information about the event involving firefighters in Daejeon. It reiterates that a fire occurred in a bathroom, resulting in the deaths of 7 people and injuries to 46 others. The text is still displayed on the same news channel, with the logo in the top right corner remaining consistent. The time remains at 13:06, and the additional news tickers at the bottom are also unchanged.",
"The current frame continues to show the exterior of the modern building with reflective glass windows. A ladder is still visible on the building, suggesting ongoing efforts to access or inspect the building. The text in the image, in Korean, reads '연합뉴스TV' indicating the news channel broadcasting the footage. The scene includes a white and blue flag, possibly indicating a symbolic or ceremonial aspect, and a person in a blue uniform, likely a police officer, is seen in the foreground. The time remains at 13:06, and there is additional news ticker information at the bottom about a railway project and exchange rates, though the focus of the video is on the building and the incident."
] |
5xiV1Z2MDX4
|
A news report is being broadcast about a fire that occurred in a bathroom of a modern building in Daejeon, resulting in 7 deaths and 46 injuries. Firefighters are involved in the response to the incident. The footage shows the exterior of the building with reflective glass windows, and a ladder is visible, indicating ongoing efforts to access or inspect the building. A police officer is present in the foreground. The news channel, 연합뉴스TV, is broadcasting the footage at 13:06, with additional news tickers providing other updates. The scene suggests an ongoing emergency response situation with a focus on the building where the tragic incident occurred.
|
Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.
|
[
"Strong winds from typhoon knocking down trees and power lines in Japan, South Korea, Taiwan, Philippines, and Russia",
"Heavy rainfall from typhoons causing flooding in Taiwan, the Philippines, and other affected East Asian countries such as Japan, South Korea, and possibly Russia",
"Typhoon storm surges hitting coastal communities in Russia's eastern regions and other affected areas such as Japan, South Korea, Taiwan, and the Philippines",
"Residents evacuating to shelters and safe zones in Japan, South Korea, Taiwan, Philippines, and Russia during typhoon events",
"Debris and damage to buildings and infrastructure in Japan, South Korea, Taiwan, Philippines, and Russia in the aftermath of the typhoon"
] |
[
"Strong winds from typhoons knocking down trees and power lines in Japan, South Korea, Taiwan, Philippines, and Russia",
"Flooding in coastal cities and low-lying areas due to typhoons in Japan, South Korea, Taiwan, Philippines, and Russia",
"People evacuating to emergency shelters during typhoon in Japan, South Korea, Taiwan, Philippines, and Russia",
"Debris-filled streets and damaged buildings after typhoon in Japan, South Korea, Taiwan, Philippines, and Russia",
"Emergency responders rescuing people from flooded areas during typhoon in Japan, South Korea, Taiwan, Philippines, and Russia"
] |
[
"Typhoon causing strong winds and knocking down trees and power lines in Japan and South Korea",
"Typhoon causing heavy rainfall and flooding in the Philippines and Taiwan",
"Typhoon storm surges hitting coastal communities in Russia and other affected areas like Japan, South Korea, Taiwan, and the Philippines",
"Residents evacuating to emergency shelters in Japan, South Korea, Taiwan, Philippines, and Russia due to typhoon",
"Debris and damaged buildings in Japan, South Korea, Taiwan, Philippines, and Russia in the aftermath of the typhoon"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%ED%9E%8C%EB%82%A8%EB%85%B8",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Hinnamnor, it makes sense to extract events that are commonly associated with a powerful tropical cyclone, such as strong winds, heavy rainfall, and storm surges, as these events are likely to be visualized in a video. The events extracted should be specific to the regions affected, including Japan, South Korea, Taiwan, the Philippines, and Russia, taking into account the typical consequences of a typhoon in these areas.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines in Japan and South Korea\n2. Heavy rainfall causing flooding in the Philippines and Taiwan\n3. Storm surges hitting coastal communities in Russia\n4. Residents evacuating to emergency shelters in affected areas\n5. Debris and damaged buildings in the aftermath of the typhoon",
"en_article_excerpt": "Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Hinnamnor",
"event_category": "disasters",
"event_name": "typhoon_hinnamno",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon, such as strong winds, heavy rainfall, and storm surges, as these are likely to be visualized in a video associated with the query. The events should be specific to the affected regions, considering the geographical characteristics and common typhoon impacts in those areas.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines in Japan and South Korea\n2. Heavy rainfall causing flooding in Taiwan and the Philippines\n3. Storm surges hitting coastal communities in Russia's eastern regions\n4. Residents evacuating to shelters and safe zones in affected areas\n5. Debris and damage to buildings and infrastructure in the aftermath of the typhoon",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the impact of a powerful typhoon, such as damage to infrastructure, evacuations, and extreme weather conditions, as these are likely to be visualized in a video associated with the query. The events should be specific to the regions affected by the typhoon, including Japan, South Korea, Taiwan, the Philippines, and Russia.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. People evacuating to emergency shelters\n4. Debris-filled streets and damaged buildings\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://twitter.com/i/status/1566887846916792320",
"video_description": "이게맞나.. #태풍 #힌남노 #포항 #역대급태풍",
"video_id": "1566887846916792320"
}
| 4
|
[
"The video captures a scene of a stormy night, with rain and strong winds. The trees in the foreground are swaying violently, indicating the intensity of the wind. A street lamp and some lights from nearby buildings provide a faint illumination, partially revealing the environment. The overall atmosphere is dark and gloomy, with rain droplets visible in the air. There are no people or vehicles visible in the frame.",
"The scene continues to depict a stormy night, with the rain and wind remaining intense. The trees in the foreground are still swaying violently, highlighting the ongoing strong winds. The street lamp and lights from nearby buildings continue to faintly illuminate the environment, maintaining the dark and gloomy atmosphere. Rain droplets are still visible in the air. A car with its rear lights on is now visible on the street, indicating some activity despite the storm. No other people or vehicles are present in the frame.",
"The scene remains consistent with the previous frames, depicting a stormy night with heavy rain and strong winds. The trees in the foreground continue to sway violently, emphasizing the storm's intensity. The street lamp and lights from nearby buildings still faintly illuminate the environment, keeping the dark and gloomy atmosphere. Rain droplets are still visible in the air. The car with its rear lights on remains present on the street, indicating some activity despite the storm. No other people or vehicles are visible in the frame.",
"The scene continues to depict a stormy night with heavy rain and strong winds. The trees in the foreground sway violently, emphasizing the storm's intensity. The street lamp and lights from nearby buildings faintly illuminate the environment, maintaining the dark and gloomy atmosphere. Rain droplets are visible in the air. The car with its rear lights on remains present on the street, indicating some activity despite the storm. No other people or vehicles are visible in the frame."
] |
1566887846916792320
|
The video depicts a stormy night scene with heavy rain and strong winds. The environment is dark and gloomy, illuminated only by a street lamp and lights from nearby buildings. Throughout the video, trees in the foreground sway violently, indicating the intensity of the wind. A car with its rear lights on is present on the street, suggesting some activity despite the severe weather conditions. The storm continues unabated throughout the video, with no other people or vehicles visible. The overall context suggests a deserted urban or suburban setting during a severe weather event.
|
Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.
|
[
"Strong winds from typhoon knocking down trees and power lines in Japan, South Korea, Taiwan, Philippines, and Russia",
"Heavy rainfall from typhoons causing flooding in Taiwan, the Philippines, and other affected East Asian countries such as Japan, South Korea, and possibly Russia",
"Typhoon storm surges hitting coastal communities in Russia's eastern regions and other affected areas such as Japan, South Korea, Taiwan, and the Philippines",
"Residents evacuating to shelters and safe zones in Japan, South Korea, Taiwan, Philippines, and Russia during typhoon events",
"Debris and damage to buildings and infrastructure in Japan, South Korea, Taiwan, Philippines, and Russia in the aftermath of the typhoon"
] |
[
"Strong winds from typhoons knocking down trees and power lines in Japan, South Korea, Taiwan, Philippines, and Russia",
"Flooding in coastal cities and low-lying areas due to typhoons in Japan, South Korea, Taiwan, Philippines, and Russia",
"People evacuating to emergency shelters during typhoon in Japan, South Korea, Taiwan, Philippines, and Russia",
"Debris-filled streets and damaged buildings after typhoon in Japan, South Korea, Taiwan, Philippines, and Russia",
"Emergency responders rescuing people from flooded areas during typhoon in Japan, South Korea, Taiwan, Philippines, and Russia"
] |
[
"Typhoon causing strong winds and knocking down trees and power lines in Japan and South Korea",
"Typhoon causing heavy rainfall and flooding in the Philippines and Taiwan",
"Typhoon storm surges hitting coastal communities in Russia and other affected areas like Japan, South Korea, Taiwan, and the Philippines",
"Residents evacuating to emergency shelters in Japan, South Korea, Taiwan, Philippines, and Russia due to typhoon",
"Debris and damaged buildings in Japan, South Korea, Taiwan, Philippines, and Russia in the aftermath of the typhoon"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%ED%9E%8C%EB%82%A8%EB%85%B8",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Hinnamnor, it makes sense to extract events that are commonly associated with a powerful tropical cyclone, such as strong winds, heavy rainfall, and storm surges, as these events are likely to be visualized in a video. The events extracted should be specific to the regions affected, including Japan, South Korea, Taiwan, the Philippines, and Russia, taking into account the typical consequences of a typhoon in these areas.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines in Japan and South Korea\n2. Heavy rainfall causing flooding in the Philippines and Taiwan\n3. Storm surges hitting coastal communities in Russia\n4. Residents evacuating to emergency shelters in affected areas\n5. Debris and damaged buildings in the aftermath of the typhoon",
"en_article_excerpt": "Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Hinnamnor",
"event_category": "disasters",
"event_name": "typhoon_hinnamno",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon, such as strong winds, heavy rainfall, and storm surges, as these are likely to be visualized in a video associated with the query. The events should be specific to the affected regions, considering the geographical characteristics and common typhoon impacts in those areas.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines in Japan and South Korea\n2. Heavy rainfall causing flooding in Taiwan and the Philippines\n3. Storm surges hitting coastal communities in Russia's eastern regions\n4. Residents evacuating to shelters and safe zones in affected areas\n5. Debris and damage to buildings and infrastructure in the aftermath of the typhoon",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the impact of a powerful typhoon, such as damage to infrastructure, evacuations, and extreme weather conditions, as these are likely to be visualized in a video associated with the query. The events should be specific to the regions affected by the typhoon, including Japan, South Korea, Taiwan, the Philippines, and Russia.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. People evacuating to emergency shelters\n4. Debris-filled streets and damaged buildings\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://twitter.com/i/status/1566774984525025281",
"video_description": "다들... 이번 태풍 힌남노 조심하세요... 현재 제주... 저희 집 나무가 90도로 순간적으로 휘어진 거 처음 봐요 바람 장난아니고 비도 장난아니니까 집에... 콕... 박혀있기",
"video_id": "1566774984525025281"
}
| 4
|
[
"The image shows a nighttime scene with a silhouette of trees and power lines. The sky appears dark, and there are no discernible people or activities in the frame.",
"The current frame continues to depict a nighttime scene with silhouettes of trees and power lines against the dark sky. There are still no discernible people or activities visible.",
"The current frame continues to depict a nighttime scene with silhouettes of trees and power lines against the dark sky. There are still no discernible people or activities visible.",
"The current frame continues to depict a nighttime scene with silhouettes of trees and power lines against the dark sky. There are still no discernible people or activities visible."
] |
1566774984525025281
|
The video depicts a static nighttime scene with silhouettes of trees and power lines against a dark sky, with no visible people or activities throughout its duration.
|
Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.
|
[
"Strong winds from typhoon knocking down trees and power lines in Japan, South Korea, Taiwan, Philippines, and Russia",
"Heavy rainfall from typhoons causing flooding in Taiwan, the Philippines, and other affected East Asian countries such as Japan, South Korea, and possibly Russia",
"Typhoon storm surges hitting coastal communities in Russia's eastern regions and other affected areas such as Japan, South Korea, Taiwan, and the Philippines",
"Residents evacuating to shelters and safe zones in Japan, South Korea, Taiwan, Philippines, and Russia during typhoon events",
"Debris and damage to buildings and infrastructure in Japan, South Korea, Taiwan, Philippines, and Russia in the aftermath of the typhoon"
] |
[
"Strong winds from typhoons knocking down trees and power lines in Japan, South Korea, Taiwan, Philippines, and Russia",
"Flooding in coastal cities and low-lying areas due to typhoons in Japan, South Korea, Taiwan, Philippines, and Russia",
"People evacuating to emergency shelters during typhoon in Japan, South Korea, Taiwan, Philippines, and Russia",
"Debris-filled streets and damaged buildings after typhoon in Japan, South Korea, Taiwan, Philippines, and Russia",
"Emergency responders rescuing people from flooded areas during typhoon in Japan, South Korea, Taiwan, Philippines, and Russia"
] |
[
"Typhoon causing strong winds and knocking down trees and power lines in Japan and South Korea",
"Typhoon causing heavy rainfall and flooding in the Philippines and Taiwan",
"Typhoon storm surges hitting coastal communities in Russia and other affected areas like Japan, South Korea, Taiwan, and the Philippines",
"Residents evacuating to emergency shelters in Japan, South Korea, Taiwan, Philippines, and Russia due to typhoon",
"Debris and damaged buildings in Japan, South Korea, Taiwan, Philippines, and Russia in the aftermath of the typhoon"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%ED%9E%8C%EB%82%A8%EB%85%B8",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Hinnamnor, it makes sense to extract events that are commonly associated with a powerful tropical cyclone, such as strong winds, heavy rainfall, and storm surges, as these events are likely to be visualized in a video. The events extracted should be specific to the regions affected, including Japan, South Korea, Taiwan, the Philippines, and Russia, taking into account the typical consequences of a typhoon in these areas.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines in Japan and South Korea\n2. Heavy rainfall causing flooding in the Philippines and Taiwan\n3. Storm surges hitting coastal communities in Russia\n4. Residents evacuating to emergency shelters in affected areas\n5. Debris and damaged buildings in the aftermath of the typhoon",
"en_article_excerpt": "Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Hinnamnor",
"event_category": "disasters",
"event_name": "typhoon_hinnamno",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon, such as strong winds, heavy rainfall, and storm surges, as these are likely to be visualized in a video associated with the query. The events should be specific to the affected regions, considering the geographical characteristics and common typhoon impacts in those areas.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines in Japan and South Korea\n2. Heavy rainfall causing flooding in Taiwan and the Philippines\n3. Storm surges hitting coastal communities in Russia's eastern regions\n4. Residents evacuating to shelters and safe zones in affected areas\n5. Debris and damage to buildings and infrastructure in the aftermath of the typhoon",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the impact of a powerful typhoon, such as damage to infrastructure, evacuations, and extreme weather conditions, as these are likely to be visualized in a video associated with the query. The events should be specific to the regions affected by the typhoon, including Japan, South Korea, Taiwan, the Philippines, and Russia.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. People evacuating to emergency shelters\n4. Debris-filled streets and damaged buildings\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://twitter.com/i/status/1567139282846089217",
"video_description": "제11호 태풍 힌남노로 인한 자연재해 피해가 속출했는데요. 포항 아파트 지하주차장 침수로 인한 대규모 실종자가 발생을 했습니다. 포항 우방신세계아파트라는 곳에서 침수로 인해 실종자가 발생했고 9월 6일 20시 40분경 실종자1명(39세 남성)을 구조했고, 21시 50분 실종자1명(여성) 구조.",
"video_id": "1567139282846089217"
}
| 4
|
[
"The current frame depicts a flooded area, likely a parking garage, with water covering the ground. A person wearing a helmet is navigating through the water, possibly to assist or check on the situation. Another individual is seen sitting or crouching nearby. In the background, there is a sign that reads \"OUT PARKING IN,\" indicating the location. Two people in the foreground are capturing the scene with their phones.",
"In the current frame, the flooding situation persists, with water covering the ground in what appears to be a parking garage. The person in a helmet continues navigating through the water, possibly to provide assistance or assess the situation. Another individual remains seated or crouched nearby. The sign \"OUT PARKING IN\" is still visible in the background, confirming the location. Two people in the foreground are still present, using their phones to document the scene.",
"The scene depicts a flood scenario with water covering the ground, likely within a parking garage. A person wearing a helmet is seen navigating through the water, possibly to provide assistance or assess the situation. Another individual is sitting or crouching nearby. In the foreground, two people are present, using their phones to document the scene. The sign \"OUT PARKING IN\" is visible in the background, confirming the location.",
"In the current frame, the flood scenario continues within the same parking garage. Multiple individuals, some wearing orange vests, are actively working together to manage the situation. One person is assisting another who appears to be submerged in water, indicating an emergency response. The presence of helmets suggests safety measures are in place. The environment remains cluttered with debris, emphasizing the severity of the flooding."
] |
1567139282846089217
|
A parking garage is severely flooded, with water covering the ground, and individuals are responding to the emergency. A person wearing a helmet navigates through the water to assess the situation or provide assistance, while another person is seated or crouched nearby. Bystanders document the scene with their phones. As the situation unfolds, multiple individuals, some wearing orange vests, work together to manage the flood, with one person assisting another who is submerged in water, indicating a rescue effort. The response suggests a coordinated emergency response with safety measures in place.
|
Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.
|
[
"Strong winds from typhoon knocking down trees and power lines in Japan, South Korea, Taiwan, Philippines, and Russia",
"Heavy rainfall from typhoons causing flooding in Taiwan, the Philippines, and other affected East Asian countries such as Japan, South Korea, and possibly Russia",
"Typhoon storm surges hitting coastal communities in Russia's eastern regions and other affected areas such as Japan, South Korea, Taiwan, and the Philippines",
"Residents evacuating to shelters and safe zones in Japan, South Korea, Taiwan, Philippines, and Russia during typhoon events",
"Debris and damage to buildings and infrastructure in Japan, South Korea, Taiwan, Philippines, and Russia in the aftermath of the typhoon"
] |
[
"Strong winds from typhoons knocking down trees and power lines in Japan, South Korea, Taiwan, Philippines, and Russia",
"Flooding in coastal cities and low-lying areas due to typhoons in Japan, South Korea, Taiwan, Philippines, and Russia",
"People evacuating to emergency shelters during typhoon in Japan, South Korea, Taiwan, Philippines, and Russia",
"Debris-filled streets and damaged buildings after typhoon in Japan, South Korea, Taiwan, Philippines, and Russia",
"Emergency responders rescuing people from flooded areas during typhoon in Japan, South Korea, Taiwan, Philippines, and Russia"
] |
[
"Typhoon causing strong winds and knocking down trees and power lines in Japan and South Korea",
"Typhoon causing heavy rainfall and flooding in the Philippines and Taiwan",
"Typhoon storm surges hitting coastal communities in Russia and other affected areas like Japan, South Korea, Taiwan, and the Philippines",
"Residents evacuating to emergency shelters in Japan, South Korea, Taiwan, Philippines, and Russia due to typhoon",
"Debris and damaged buildings in Japan, South Korea, Taiwan, Philippines, and Russia in the aftermath of the typhoon"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%ED%9E%8C%EB%82%A8%EB%85%B8",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Hinnamnor, it makes sense to extract events that are commonly associated with a powerful tropical cyclone, such as strong winds, heavy rainfall, and storm surges, as these events are likely to be visualized in a video. The events extracted should be specific to the regions affected, including Japan, South Korea, Taiwan, the Philippines, and Russia, taking into account the typical consequences of a typhoon in these areas.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines in Japan and South Korea\n2. Heavy rainfall causing flooding in the Philippines and Taiwan\n3. Storm surges hitting coastal communities in Russia\n4. Residents evacuating to emergency shelters in affected areas\n5. Debris and damaged buildings in the aftermath of the typhoon",
"en_article_excerpt": "Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Hinnamnor",
"event_category": "disasters",
"event_name": "typhoon_hinnamno",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon, such as strong winds, heavy rainfall, and storm surges, as these are likely to be visualized in a video associated with the query. The events should be specific to the affected regions, considering the geographical characteristics and common typhoon impacts in those areas.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines in Japan and South Korea\n2. Heavy rainfall causing flooding in Taiwan and the Philippines\n3. Storm surges hitting coastal communities in Russia's eastern regions\n4. Residents evacuating to shelters and safe zones in affected areas\n5. Debris and damage to buildings and infrastructure in the aftermath of the typhoon",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the impact of a powerful typhoon, such as damage to infrastructure, evacuations, and extreme weather conditions, as these are likely to be visualized in a video associated with the query. The events should be specific to the regions affected by the typhoon, including Japan, South Korea, Taiwan, the Philippines, and Russia.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. People evacuating to emergency shelters\n4. Debris-filled streets and damaged buildings\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://twitter.com/i/status/1566015450982494209",
"video_description": "얘들아… 초속 54미터 이상이면 건물도 붕괴될 정도의 바람이라는데 이번 힌남노는 매미보다 중심 기압이 낮은 더 강력한 태풍이라 제주/남부해안지역엔 초속 60미터까지도 예상된대.. 영동 쪽은 초속 3~40미터 수도권은 초속 15미터 예상… 다들 진짜 조심하자 대비할 수 있는 부분은 꼭 하고ㅜ",
"video_id": "1566015450982494209"
}
| 4
|
[
"The current frame shows a weather forecast presentation. A map displaying wind speeds in different regions is shown on the left, with areas colored in yellow, green, blue, and purple, indicating varying wind speeds from 50-60 m/s to 15 m/s. A man is speaking on the right side of the frame, likely providing commentary on the weather conditions. The text at the bottom reads \"서울 등 수도권에도 초속 15m 돌풍 불듯...시설물 피해 우려,\" which translates to \"Severe gusts of up to 15 meters per second expected in Seoul and the metropolitan area...Concerns about damage to infrastructure.\"",
"The video is a weather forecast presentation. The map on the left displays wind speeds in different regions, with areas colored in yellow, green, blue, and purple, indicating varying wind speeds from 50-60 m/s to 15 m/s. A man is speaking on the right side of the frame, providing commentary on the weather conditions. The text at the bottom reads \"2003년 태풍 '매미'로 폭풍해일 발생...인명 피해 발생,\" which translates to \"2003 typhoon 'Maeby' caused a storm surge...causing casualties.\"",
"The video continues the weather forecast presentation. On the left side, a detailed map illustrates wind speeds across different regions, with areas marked in varying colors such as yellow, green, blue, and purple, indicating wind speeds ranging from 50-60 m/s to 15 m/s. The man on the right, dressed in a suit and tie, is providing commentary on the weather conditions. The text at the bottom of the screen reads \"2003년 태풍 '매미'로 폭풍해일 발생...인명 피해 발생,\" which translates to \"2003 typhoon 'Maeby' caused a storm surge...causing casualties.\"",
"The video continues with the man on the right, dressed in a suit and tie, continuing his commentary on the weather conditions. The left side of the screen now shows a nighttime scene of a street with illuminated streetlights and a red traffic signal at a distance, suggesting a discussion on the impact of weather on urban areas. The text at the bottom of the screen remains \"2003년 태풍 '매미'로 폭풍해일 발생...인명 피해 발생,\" which translates to \"2003 typhoon 'Maeby' caused a storm surge...causing casualties.\" The overall setting indicates a weather forecast presentation focusing on past weather events and their effects."
] |
1566015450982494209
|
A weather forecast presentation is being given by a man, discussing severe wind speeds and past weather events, including the 2003 typhoon 'Maeby' that caused a storm surge and casualties. The presentation displays maps of wind speeds in different regions and shows footage of urban areas at night, highlighting the potential impact of weather on infrastructure and daily life. The forecast warns of strong gusts in Seoul and the metropolitan area, raising concerns about damage to infrastructure.
|
Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.
|
[
"Strong winds from typhoon knocking down trees and power lines in Japan, South Korea, Taiwan, Philippines, and Russia",
"Heavy rainfall from typhoons causing flooding in Taiwan, the Philippines, and other affected East Asian countries such as Japan, South Korea, and possibly Russia",
"Typhoon storm surges hitting coastal communities in Russia's eastern regions and other affected areas such as Japan, South Korea, Taiwan, and the Philippines",
"Residents evacuating to shelters and safe zones in Japan, South Korea, Taiwan, Philippines, and Russia during typhoon events",
"Debris and damage to buildings and infrastructure in Japan, South Korea, Taiwan, Philippines, and Russia in the aftermath of the typhoon"
] |
[
"Strong winds from typhoons knocking down trees and power lines in Japan, South Korea, Taiwan, Philippines, and Russia",
"Flooding in coastal cities and low-lying areas due to typhoons in Japan, South Korea, Taiwan, Philippines, and Russia",
"People evacuating to emergency shelters during typhoon in Japan, South Korea, Taiwan, Philippines, and Russia",
"Debris-filled streets and damaged buildings after typhoon in Japan, South Korea, Taiwan, Philippines, and Russia",
"Emergency responders rescuing people from flooded areas during typhoon in Japan, South Korea, Taiwan, Philippines, and Russia"
] |
[
"Typhoon causing strong winds and knocking down trees and power lines in Japan and South Korea",
"Typhoon causing heavy rainfall and flooding in the Philippines and Taiwan",
"Typhoon storm surges hitting coastal communities in Russia and other affected areas like Japan, South Korea, Taiwan, and the Philippines",
"Residents evacuating to emergency shelters in Japan, South Korea, Taiwan, Philippines, and Russia due to typhoon",
"Debris and damaged buildings in Japan, South Korea, Taiwan, Philippines, and Russia in the aftermath of the typhoon"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%ED%9E%8C%EB%82%A8%EB%85%B8",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Hinnamnor, it makes sense to extract events that are commonly associated with a powerful tropical cyclone, such as strong winds, heavy rainfall, and storm surges, as these events are likely to be visualized in a video. The events extracted should be specific to the regions affected, including Japan, South Korea, Taiwan, the Philippines, and Russia, taking into account the typical consequences of a typhoon in these areas.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines in Japan and South Korea\n2. Heavy rainfall causing flooding in the Philippines and Taiwan\n3. Storm surges hitting coastal communities in Russia\n4. Residents evacuating to emergency shelters in affected areas\n5. Debris and damaged buildings in the aftermath of the typhoon",
"en_article_excerpt": "Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Hinnamnor",
"event_category": "disasters",
"event_name": "typhoon_hinnamno",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typical of a powerful typhoon, such as strong winds, heavy rainfall, and storm surges, as these are likely to be visualized in a video associated with the query. The events should be specific to the affected regions, considering the geographical characteristics and common typhoon impacts in those areas.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines in Japan and South Korea\n2. Heavy rainfall causing flooding in Taiwan and the Philippines\n3. Storm surges hitting coastal communities in Russia's eastern regions\n4. Residents evacuating to shelters and safe zones in affected areas\n5. Debris and damage to buildings and infrastructure in the aftermath of the typhoon",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Hinnamnor, known in the Philippines as Super Typhoon Henry, was a very large and powerful tropical cyclone that impacted Japan, South Korea, Taiwan, the Philippines, and Russia.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events related to the impact of a powerful typhoon, such as damage to infrastructure, evacuations, and extreme weather conditions, as these are likely to be visualized in a video associated with the query. The events should be specific to the regions affected by the typhoon, including Japan, South Korea, Taiwan, the Philippines, and Russia.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal cities and low-lying areas\n3. People evacuating to emergency shelters\n4. Debris-filled streets and damaged buildings\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://twitter.com/i/status/1566948535899041792",
"video_description": "칭구가 보내준 집앞 영상 괜찮아??? 응 근데 나도 안괜찮아 다들 괜찮으시길 #태풍_힌남노_저리가",
"video_id": "1566948535899041792"
}
| 4
|
[
"The image shows a large landslide with a significant portion of earth and debris having collapsed down a hillside. A modern building is situated near the bottom of the slope, partially buried by the landslide. The area around the building is covered with mud and rocks. In the background, several high-rise apartment buildings are visible, suggesting an urban environment. Trees line the hillside, and a vehicle can be seen parked nearby. The sky appears overcast.",
"The current frame depicts a significant landslide that has occurred, with a substantial amount of earth and debris having collapsed down a hillside. The modern building at the bottom of the slope is partially buried by the landslide, and the area around it is covered with mud and rocks. Trees line the hillside, and a vehicle can be seen parked nearby. The high-rise apartment buildings in the background indicate an urban setting, and the overcast sky suggests the landslide may have been triggered by heavy rain or other weather conditions.",
"The current frame depicts the aftermath of a significant landslide, with a substantial amount of earth and debris having collapsed down a hillside, partially burying a modern building at the bottom of the slope. The area around the building is covered with mud and rocks, and trees line the hillside. A vehicle remains parked nearby, indicating that the landslide was sudden and unexpected. The high-rise apartment buildings in the background continue to indicate an urban setting, and the overcast sky suggests that the landslide may have been triggered by heavy rain or other weather conditions. No visible text is present to provide additional context.",
"The current frame continues to show the aftermath of the landslide, with the modern building at the bottom of the slope still partially buried under the mud and rocks. The area around the building remains covered with debris, and the trees lining the hillside are still visible. The vehicle remains parked nearby, reinforcing the suddenness of the event. The high-rise apartment buildings in the background remain consistent with the previous frame, maintaining the urban setting. The overcast sky persists, suggesting ongoing or recent heavy weather conditions that may have contributed to the landslide. There is no visible text in this frame to offer additional context."
] |
1566948535899041792
|
A significant landslide occurs in an urban area, causing a substantial amount of earth and debris to collapse down a hillside and partially bury a modern building at the bottom of the slope. The area around the building is covered with mud and rocks, and trees line the hillside. A vehicle parked nearby suggests the landslide was sudden and unexpected. The presence of high-rise apartment buildings in the background and an overcast sky indicates the event took place in an urban environment, possibly triggered by heavy rain or other weather conditions. The landslide appears to have been a singular, dramatic event, with no visible signs of ongoing activity or response in the provided frames.
|
Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines, China, and Korea in 2019",
"Heavy rainfall and flooding from Typhoon in Philippines, China, and Korea in 2019",
"Storm surges causing coastal erosion and damage during Typhoon in the Philippines, China, and Korea in 2019",
"People evacuating to safer areas or shelters during the 2019 typhoon in the Philippines, China, and Korea",
"Debris and destroyed buildings scattered in the aftermath of the 2019 typhoon in the Philippines, China, and Korea"
] |
[
"Buildings and houses destroyed by strong winds from Typhoon in Philippines, China, and Korea in 2019",
"Flooding in coastal areas and low-lying regions during Typhoon in the Philippines, China, and Korea in 2019",
"People evacuating to higher ground or shelters during the 2019 typhoon in the Philippines, China, and Korea",
"Trees and power lines uprooted or knocked down during Typhoon in Philippines, China, and Korea in 2019",
"Rescue teams providing aid to affected communities during the 2019 typhoon in the Philippines, China, and Korea"
] |
[
"Strong winds from Typhoon knocking down trees and power lines in the Philippines, China, and Korea in 2019",
"Heavy rainfall and flooding from Typhoon in Philippines, China, and Korea in 2019",
"Storm surges caused by typhoons in the Philippines, China, and Korea in 2019 resulting in coastal erosion and damage",
"People evacuating to safer areas during Typhoon in the Philippines, China, and Korea in 2019",
"Debris and wreckage scattered on roads and buildings after Typhoon in Philippines, China, and Korea in 2019"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%A7%81%EB%A7%81_(2019%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Lingling, it makes sense to extract events that are commonly associated with a powerful typhoon, such as strong winds, heavy rainfall, and storm surges, which could be visualized in a video. These events are likely to have occurred in the affected countries, including the Philippines, China, and Korea, and can be reasonably captioned in a video frame.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Storm surges causing coastal erosion and damage\n4. People evacuating to safer areas\n5. Debris and wreckage scattered on roads and buildings",
"en_article_excerpt": "Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Lingling_(2019)",
"event_category": "disasters",
"event_name": "typhoon_lingling",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about Typhoon Lingling, it makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, heavy rainfall, and storm surges, which could be visualized in a video. These events are likely to occur in the affected countries, including the Philippines, China, and Korea, and can be reasonably captioned in a video frame.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Storm surges causing coastal erosion and damage\n4. People evacuating to safer areas or shelters\n5. Debris and destroyed buildings scattered in the typhoon's aftermath",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typically associated with the aftermath or impact of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Lingling. The events extracted should be specific to the regions affected by the typhoon, including the Philippines, China, and Korea.\n\nEVENTS:\n1. Buildings and houses destroyed by strong winds\n2. Flooding in coastal areas and low-lying regions\n3. People evacuating to higher ground or shelters\n4. Trees and power lines uprooted or knocked down\n5. Rescue teams providing aid to affected communities",
"video_URL": "http://youtube.com/watch?v=QJY4XlSNVf8",
"video_description": "#태풍 #링링 #제13호 #13호 #중국 (서울=연합뉴스) 제13호 태풍 '링링'이 세력을 키우며 한반도를 향해 북상하고 있습니다. 5일 기상청에 따르면 '링링'은 이날 오전 9시 현재 일본 오키나와 남서쪽 약 370㎞ 해상에서 시속 19㎞로 북상 중입니다. 발생 초기 강도가 '약'이던 '링링'은 수온이 높은 해역을 지나며 이날 현재 '강'으로 세력이 강해진 상태입니다. 한국과 미국, 일본 기상청 모두 '링링'이 토요일인 7일 수도권을 강타할 것으로 예보해 철저한 대비가 필요합니다. 중국 저장성과 하이난성에서는 지난 2일 이후 '링링'과 제14호 태풍 '가지키'의 영향으로 폭우 피해가 잇따랐는데요. 영상으로 보시죠. ◆ 연합뉴스 홈페이지→ http://www.yna.co.kr/ ◆ 이 시각 많이 본 기사 → https://goo.gl/VQTsSZ ◆ 오늘의 핫뉴스 → https://goo.gl/WyGXpG ◆ 현장영상 → https://goo.gl/5aZcx8 ◆ 카드뉴스 →https://goo.gl/QKfDTH ◆연합뉴스 공식 SNS◆ ◇페이스북→ https://www.facebook.com/yonhap/ ▣ 연합뉴스 유튜브 채널 구독 : https://goo.gl/pL7TmT ▣ 연합뉴스 인스타 : https://goo.gl/UbqiQb",
"video_id": "QJY4XlSNVf8"
}
| 4
|
[
"The current frame depicts a rainy scene with a backdrop of dense green trees, indicating a forest or wooded area. The text on the image reads \"연합뉴스\" (Yonhap News) and \"지난 2일 이후\" (After the 2nd), suggesting that the video is likely discussing weather conditions following a specific date.",
"The current frame continues to show a rainy scene with dense green trees in the background, maintaining the forested area setting. The text on the image reads \"연합뉴스\" (Yonhap News) and \"림림의 이동 속도가 점점 빨라지면서\" (As the speed of the rain intensifies), indicating an increase in the intensity of rainfall, possibly due to the storm or heavy rain discussed in the previous frame.",
"The current frame shows a stormy scene with heavy rain and mist obscuring the view of the area. The text on the image reads \"연합뉴스\" (Yonhap News) and \"림림이 6~7일 우리나라 전역에 영향을 미칠 것으로 예상됨에 따라\" (As it is expected to affect the entire country on the 6th and 7th), indicating that the storm is expected to impact the entire country within the next few days.",
"The current frame continues to depict a stormy scene with heavy rain and mist obscuring the view of the area. The text on the image reads \"림림이 6~7일 우리나라 전역에 영향을 미칠 것으로 예상됨에 따라\" (As it is expected to affect the entire country on the 6th and 7th), reiterating the expectation of significant impact from the storm in the coming days."
] |
QJY4XlSNVf8
|
A news report from Yonhap News discusses a severe weather condition, specifically a storm with heavy rain, expected to affect the entire country on the 6th and 7th. The video shows a forested area being hit by intense rainfall, with the storm's speed and impact intensifying over time, causing reduced visibility due to heavy rain and mist. The report appears to be updating viewers on the storm's progression and its anticipated effects on the country.
|
Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines, China, and Korea in 2019",
"Heavy rainfall and flooding from Typhoon in Philippines, China, and Korea in 2019",
"Storm surges causing coastal erosion and damage during Typhoon in the Philippines, China, and Korea in 2019",
"People evacuating to safer areas or shelters during the 2019 typhoon in the Philippines, China, and Korea",
"Debris and destroyed buildings scattered in the aftermath of the 2019 typhoon in the Philippines, China, and Korea"
] |
[
"Buildings and houses destroyed by strong winds from Typhoon in Philippines, China, and Korea in 2019",
"Flooding in coastal areas and low-lying regions during Typhoon in the Philippines, China, and Korea in 2019",
"People evacuating to higher ground or shelters during the 2019 typhoon in the Philippines, China, and Korea",
"Trees and power lines uprooted or knocked down during Typhoon in Philippines, China, and Korea in 2019",
"Rescue teams providing aid to affected communities during the 2019 typhoon in the Philippines, China, and Korea"
] |
[
"Strong winds from Typhoon knocking down trees and power lines in the Philippines, China, and Korea in 2019",
"Heavy rainfall and flooding from Typhoon in Philippines, China, and Korea in 2019",
"Storm surges caused by typhoons in the Philippines, China, and Korea in 2019 resulting in coastal erosion and damage",
"People evacuating to safer areas during Typhoon in the Philippines, China, and Korea in 2019",
"Debris and wreckage scattered on roads and buildings after Typhoon in Philippines, China, and Korea in 2019"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%A7%81%EB%A7%81_(2019%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Lingling, it makes sense to extract events that are commonly associated with a powerful typhoon, such as strong winds, heavy rainfall, and storm surges, which could be visualized in a video. These events are likely to have occurred in the affected countries, including the Philippines, China, and Korea, and can be reasonably captioned in a video frame.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Storm surges causing coastal erosion and damage\n4. People evacuating to safer areas\n5. Debris and wreckage scattered on roads and buildings",
"en_article_excerpt": "Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Lingling_(2019)",
"event_category": "disasters",
"event_name": "typhoon_lingling",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about Typhoon Lingling, it makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, heavy rainfall, and storm surges, which could be visualized in a video. These events are likely to occur in the affected countries, including the Philippines, China, and Korea, and can be reasonably captioned in a video frame.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Storm surges causing coastal erosion and damage\n4. People evacuating to safer areas or shelters\n5. Debris and destroyed buildings scattered in the typhoon's aftermath",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typically associated with the aftermath or impact of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Lingling. The events extracted should be specific to the regions affected by the typhoon, including the Philippines, China, and Korea.\n\nEVENTS:\n1. Buildings and houses destroyed by strong winds\n2. Flooding in coastal areas and low-lying regions\n3. People evacuating to higher ground or shelters\n4. Trees and power lines uprooted or knocked down\n5. Rescue teams providing aid to affected communities",
"video_URL": "http://youtube.com/watch?v=ZFR-SBQwtTw",
"video_description": "태풍 '링링' 발생…주후반 한반도 영향 가능성 오늘 오전 9시쯤 필리핀 동쪽 해상에서 13호 태풍 '링링'이 발생했습니다. 강풍반경 200km의 소형 태풍인 '링링'은 계속 북상해서 수요일쯤 타이완 동쪽 해상에 접근하겠습니다. 태풍 '링링'은 금요일 오전쯤 제주도 서귀포 남서쪽 390km 해상까지 진출하겠고 이후 서해상을 향할 것으로 보입니다. 기상청은 태풍 발생 초기라 경로가 유동적이지만 주 후반부터 한반도 부근으로 북상할 가능성이 크다고 내다봤습니다. 연합뉴스TV 기사문의 및 제보 : 카톡/라인 jebo23 ▣ 연합뉴스TV 유튜브 채널 구독 https://goo.gl/VuCJMi ▣ 대한민국 뉴스의 시작 연합뉴스TV / Yonhap News TV http://www.yonhapnewstv.co.kr/",
"video_id": "ZFR-SBQwtTw"
}
| 4
|
[
"The video shows a turbulent sea with waves crashing onto the shore. The text overlay in Korean indicates a typhoon named 'Liming' has formed, and there is a possibility of significant impact after the weekend.",
"The video captures a bustling street scene amidst the aftermath of Typhoon 'Liming.' Two men are seen riding on the back of a white truck labeled \"CITY ENGINEERS OFFICE,\" possibly assessing damage or mobilizing resources. The street is lined with palm trees, and in the background, another blue truck is visible. On the left, makeshift shelters made of tarpaulin suggest areas of disarray and recovery efforts. The overlay text in Korean reiterates the formation of Typhoon 'Liming' and the potential for continued impact, emphasizing the ongoing challenges faced by the city's infrastructure and residents.",
"The video shows a coastal area affected by Typhoon 'Liming,' with waves crashing against a breakwater, indicating the typhoon's impact on the local environment. A person is seen walking on a wet surface, likely assessing the damage or monitoring the situation. The text in Korean at the bottom of the frame reiterates the formation of Typhoon 'Liming' and the potential for continued impact, emphasizing the ongoing challenges faced by the coastal area and its residents.",
"The current frame shows a boat docked at a harbor with multiple boats visible, indicating a return to normalcy or efforts to resume activities following the typhoon. The presence of multiple boats suggests a fishing or coastal community preparing for operations or assessing damages. The text in Korean at the top of the frame mentions \"연합뉴스 TV\" and the time \"13:21,\" indicating a news report from a news channel, possibly providing an update on the situation or progress made after the typhoon. The text at the bottom of the frame further details financial information related to a government budget, which may be related to relief or recovery efforts in the aftermath of Typhoon 'Liming.'"
] |
ZFR-SBQwtTw
|
A typhoon named 'Liming' has formed and is expected to have a significant impact on a coastal area. The video depicts the aftermath of the typhoon, showing a turbulent sea, damaged infrastructure, and recovery efforts. City engineers are assessing the damage and mobilizing resources, while residents are dealing with the challenges posed by the storm. The typhoon has affected the local environment, with waves crashing against the breakwater, and people are working to monitor the situation and resume normal activities. As the community begins to recover, fishing or coastal operations are preparing to resume, and news reports are providing updates on the situation and government relief efforts, including budget allocations for recovery.
|
Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines, China, and Korea in 2019",
"Heavy rainfall and flooding from Typhoon in Philippines, China, and Korea in 2019",
"Storm surges causing coastal erosion and damage during Typhoon in the Philippines, China, and Korea in 2019",
"People evacuating to safer areas or shelters during the 2019 typhoon in the Philippines, China, and Korea",
"Debris and destroyed buildings scattered in the aftermath of the 2019 typhoon in the Philippines, China, and Korea"
] |
[
"Buildings and houses destroyed by strong winds from Typhoon in Philippines, China, and Korea in 2019",
"Flooding in coastal areas and low-lying regions during Typhoon in the Philippines, China, and Korea in 2019",
"People evacuating to higher ground or shelters during the 2019 typhoon in the Philippines, China, and Korea",
"Trees and power lines uprooted or knocked down during Typhoon in Philippines, China, and Korea in 2019",
"Rescue teams providing aid to affected communities during the 2019 typhoon in the Philippines, China, and Korea"
] |
[
"Strong winds from Typhoon knocking down trees and power lines in the Philippines, China, and Korea in 2019",
"Heavy rainfall and flooding from Typhoon in Philippines, China, and Korea in 2019",
"Storm surges caused by typhoons in the Philippines, China, and Korea in 2019 resulting in coastal erosion and damage",
"People evacuating to safer areas during Typhoon in the Philippines, China, and Korea in 2019",
"Debris and wreckage scattered on roads and buildings after Typhoon in Philippines, China, and Korea in 2019"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%A7%81%EB%A7%81_(2019%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Lingling, it makes sense to extract events that are commonly associated with a powerful typhoon, such as strong winds, heavy rainfall, and storm surges, which could be visualized in a video. These events are likely to have occurred in the affected countries, including the Philippines, China, and Korea, and can be reasonably captioned in a video frame.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Storm surges causing coastal erosion and damage\n4. People evacuating to safer areas\n5. Debris and wreckage scattered on roads and buildings",
"en_article_excerpt": "Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Lingling_(2019)",
"event_category": "disasters",
"event_name": "typhoon_lingling",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about Typhoon Lingling, it makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, heavy rainfall, and storm surges, which could be visualized in a video. These events are likely to occur in the affected countries, including the Philippines, China, and Korea, and can be reasonably captioned in a video frame.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Storm surges causing coastal erosion and damage\n4. People evacuating to safer areas or shelters\n5. Debris and destroyed buildings scattered in the typhoon's aftermath",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typically associated with the aftermath or impact of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Lingling. The events extracted should be specific to the regions affected by the typhoon, including the Philippines, China, and Korea.\n\nEVENTS:\n1. Buildings and houses destroyed by strong winds\n2. Flooding in coastal areas and low-lying regions\n3. People evacuating to higher ground or shelters\n4. Trees and power lines uprooted or knocked down\n5. Rescue teams providing aid to affected communities",
"video_URL": "http://youtube.com/watch?v=8-DXe4qYo2E",
"video_description": "#태풍 #링링 (서울=연합뉴스) 제13호 태풍 '링링'이 필리핀 동부에서 발생했습니다. 이 태풍의 중심기압은 1천hPa(헥토파스칼), 중심 부근 최대 풍속은 초속 18ｍ(시속 65㎞)이며, 강풍 반경은 200㎞입니다. 한반도에 접근할 무렵 지금보다 크고 강해질 것으로 전망돼 대비가 필요한데요. 자세한 내용은 영상으로 보시죠. ◆ 연합뉴스 홈페이지→ http://www.yna.co.kr/ ◆ 이 시각 많이 본 기사 → https://goo.gl/VQTsSZ ◆ 오늘의 핫뉴스 → https://goo.gl/WyGXpG ◆ 현장영상 → https://goo.gl/5aZcx8 ◆ 카드뉴스 →https://goo.gl/QKfDTH ◆연합뉴스 공식 SNS◆ ◇페이스북→ https://www.facebook.com/yonhap/ ▣ 연합뉴스 유튜브 채널 구독 : https://goo.gl/pL7TmT ▣ 연합뉴스 인스타 : https://goo.gl/UbqiQb",
"video_id": "8-DXe4qYo2E"
}
| 4
|
[
"The video frame depicts a turbulent ocean scene with large waves crashing against rocks. A red lighthouse stands prominently amidst the stormy weather. The Korean text on the frame mentions a typhoon affecting a 560km area along the eastern coast.",
"The video frame shows a view of a turbulent ocean with waves crashing against a concrete barrier. The Korean text on the frame states that the typhoon has reached the area by 9 a.m. on the 6th, affecting a region approximately 390km southeast of the southern part of Jeju Island.",
"The image shows a foggy, rain-drenched street with low visibility, indicating severe weather conditions. The Korean text in the image suggests that the storm named \"Ringming\" has caused significant flooding in regions near Daeban and Zhonggu Shanghuai. It highlights the risk of flooding in areas downstream, particularly in the southern region, due to the storm's impact. The text appears to be a news report, providing a warning or update about the storm's effects.",
"The current frame features a cartoon illustration of a man with glasses, pointing upwards with a smile on his face. He is associated with a speech bubble that contains Korean text. Above him, there is a larger speech bubble with more Korean text and a logo for the Yonhap News Agency. The illustration suggests a humorous or lighthearted tone, possibly as a contrast to the serious weather conditions described in the previous frame. The logo indicates that the illustration is part of a news report, possibly offering a lighter moment within the coverage of the storm named \"Ringming.\""
] |
8-DXe4qYo2E
|
A typhoon, identified as "Ringming," is affecting a large area along the eastern coast, causing turbulent ocean scenes with large waves crashing against rocks and concrete barriers. The storm has reached the southern part of Jeju Island by 9 a.m. on the 6th, bringing severe weather conditions including heavy rain and fog, resulting in significant flooding in regions near Daeban and Zhonggu Shanghuai. News reports, including those from the Yonhap News Agency, are providing updates and warnings about the storm's impact, with a brief moment of levity introduced through a cartoon illustration. Overall, the video depicts the storm's effects on the environment and the efforts to inform the public about the situation.
|
Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines, China, and Korea in 2019",
"Heavy rainfall and flooding from Typhoon in Philippines, China, and Korea in 2019",
"Storm surges causing coastal erosion and damage during Typhoon in the Philippines, China, and Korea in 2019",
"People evacuating to safer areas or shelters during the 2019 typhoon in the Philippines, China, and Korea",
"Debris and destroyed buildings scattered in the aftermath of the 2019 typhoon in the Philippines, China, and Korea"
] |
[
"Buildings and houses destroyed by strong winds from Typhoon in Philippines, China, and Korea in 2019",
"Flooding in coastal areas and low-lying regions during Typhoon in the Philippines, China, and Korea in 2019",
"People evacuating to higher ground or shelters during the 2019 typhoon in the Philippines, China, and Korea",
"Trees and power lines uprooted or knocked down during Typhoon in Philippines, China, and Korea in 2019",
"Rescue teams providing aid to affected communities during the 2019 typhoon in the Philippines, China, and Korea"
] |
[
"Strong winds from Typhoon knocking down trees and power lines in the Philippines, China, and Korea in 2019",
"Heavy rainfall and flooding from Typhoon in Philippines, China, and Korea in 2019",
"Storm surges caused by typhoons in the Philippines, China, and Korea in 2019 resulting in coastal erosion and damage",
"People evacuating to safer areas during Typhoon in the Philippines, China, and Korea in 2019",
"Debris and wreckage scattered on roads and buildings after Typhoon in Philippines, China, and Korea in 2019"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%A7%81%EB%A7%81_(2019%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Lingling, it makes sense to extract events that are commonly associated with a powerful typhoon, such as strong winds, heavy rainfall, and storm surges, which could be visualized in a video. These events are likely to have occurred in the affected countries, including the Philippines, China, and Korea, and can be reasonably captioned in a video frame.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Storm surges causing coastal erosion and damage\n4. People evacuating to safer areas\n5. Debris and wreckage scattered on roads and buildings",
"en_article_excerpt": "Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Lingling_(2019)",
"event_category": "disasters",
"event_name": "typhoon_lingling",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about Typhoon Lingling, it makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, heavy rainfall, and storm surges, which could be visualized in a video. These events are likely to occur in the affected countries, including the Philippines, China, and Korea, and can be reasonably captioned in a video frame.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Storm surges causing coastal erosion and damage\n4. People evacuating to safer areas or shelters\n5. Debris and destroyed buildings scattered in the typhoon's aftermath",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typically associated with the aftermath or impact of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Lingling. The events extracted should be specific to the regions affected by the typhoon, including the Philippines, China, and Korea.\n\nEVENTS:\n1. Buildings and houses destroyed by strong winds\n2. Flooding in coastal areas and low-lying regions\n3. People evacuating to higher ground or shelters\n4. Trees and power lines uprooted or knocked down\n5. Rescue teams providing aid to affected communities",
"video_URL": "http://youtube.com/watch?v=taOktWIvk20",
"video_description": "#태풍 #링링 #서울태풍 #수도권태풍 #최근접 #초강력태풍 (서울=연합뉴스) 한반도를 강타할 초강력 제13호 태풍 '링링'이 왔습니다. 기상청에 따르면 태풍은 7일 새벽 서귀포 서쪽 약 150㎞ 해상을 통과, 오전 6시께 전남 목포 서쪽 약 140㎞ 해상을 지났습니다. 정오께 서울 서남서쪽 약 140㎞ 해상을 지날 것으로 보이는데요. 오후 2시께 태풍 중심이 서울에 제일 가까울 전망입니다. 영상으로 보시겠습니다. [영상 : 연합뉴스TV] ◆ 연합뉴스 홈페이지→ http://www.yna.co.kr/ ◆ 이 시각 많이 본 기사 → https://goo.gl/VQTsSZ ◆ 오늘의 핫뉴스 → https://goo.gl/WyGXpG ◆ 현장영상 → https://goo.gl/5aZcx8 ◆ 카드뉴스 →https://goo.gl/QKfDTH ◆연합뉴스 공식 SNS◆ ◇페이스북→ https://www.facebook.com/yonhap/ ▣ 연합뉴스 유튜브 채널 구독 : https://goo.gl/pL7TmT ▣ 연합뉴스 인스타 : https://goo.gl/UbqiQb",
"video_id": "taOktWIvk20"
}
| 4
|
[
"A fishing boat is docked at a pier. Several individuals are present on the boat, engaging in various activities. The environment appears overcast, suggesting cloudy weather. The boat is equipped with fishing gear, including nets and other equipment, indicating a fishing operation. The text on the image, in Korean, mentions \"연합뉴스\" and \"태풍 제주 강타...서귀포 서쪽 약 150km 해상 통과,\" which translates to \"Joint News\" and \"Typhoon hits Jeju...passes approximately 150km west of Seogwipo,\" indicating that the footage is related to the impact of a typhoon in the region.",
"The current frame shows a fishing boat with various equipment visible on its deck, docked near other boats, all likely engaged in fishing activities. The environment remains overcast, consistent with the cloudy weather described previously. Text on the image reads \"연합뉴스\" and \"정오께 서울 서남쪽 약 140km 해상 지날듯,\" translating to \"Joint News\" and \"At noon, about 140km southwest of Seoul at sea,\" suggesting the footage continues to relate to the ongoing impact of the typhoon in the region.",
"The current frame shows a close-up of a green arrow pointing to the left, likely indicating the direction of traffic flow for vehicles or pedestrians. The arrow is brightly lit, suggesting its visibility is important for navigation or safety. The environment appears to be outdoors, possibly at an intersection or a designated pedestrian crossing, given the purpose of the arrow. The sky is overcast, consistent with the cloudy weather described previously, which may affect visibility and travel conditions. The text on the image reads \"연합뉴스\" and \"태풍특보 발효.. '매우 강한 바람 유의해야,'\" translating to \"Joint News\" and \"Typhoon Special Report Issued.. 'Be Cautious of Strong Winds,'\" indicating that the footage continues to relate to the ongoing impact of the typhoon in the region. The green arrow may serve as a guide for safe navigation during the typhoon, emphasizing the need for caution due to the strong winds.",
"The current frame shows a blue background with geometric patterns. The text in the image reads \"연합뉴스\" and \"YONHAP NEWS AGENCY,\" indicating the logo and name of the news agency. There are no visible objects or actions taking place, as the frame is solely focused on the branding of the news organization. This frame likely serves as an introductory or transitional element within the video, possibly before or after the content related to the typhoon special report."
] |
taOktWIvk20
|
A video depicting the impact of a typhoon in the Jeju region and possibly the broader area including Seoul. The footage shows a fishing boat docked at a pier with individuals on board, surrounded by fishing gear, under overcast weather conditions. The boat is part of a larger fishing operation, with other boats nearby. As the typhoon progresses, the video shifts focus to a traffic or pedestrian navigation aid, a green arrow, illuminated to guide movement safely during the storm. The typhoon's strong winds are a major concern, as indicated by a special report issued by the news agency. The video is a report from YONHAP NEWS AGENCY, as evident from the logo and branding displayed in one of the frames. Overall, the video aims to inform viewers about the typhoon's effects, particularly on maritime activities and public safety, emphasizing caution due to the severe weather conditions.
|
Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.
|
[
"Strong winds from typhoons knocking down trees and power lines in the Philippines, China, and Korea in 2019",
"Heavy rainfall and flooding from Typhoon in Philippines, China, and Korea in 2019",
"Storm surges causing coastal erosion and damage during Typhoon in the Philippines, China, and Korea in 2019",
"People evacuating to safer areas or shelters during the 2019 typhoon in the Philippines, China, and Korea",
"Debris and destroyed buildings scattered in the aftermath of the 2019 typhoon in the Philippines, China, and Korea"
] |
[
"Buildings and houses destroyed by strong winds from Typhoon in Philippines, China, and Korea in 2019",
"Flooding in coastal areas and low-lying regions during Typhoon in the Philippines, China, and Korea in 2019",
"People evacuating to higher ground or shelters during the 2019 typhoon in the Philippines, China, and Korea",
"Trees and power lines uprooted or knocked down during Typhoon in Philippines, China, and Korea in 2019",
"Rescue teams providing aid to affected communities during the 2019 typhoon in the Philippines, China, and Korea"
] |
[
"Strong winds from Typhoon knocking down trees and power lines in the Philippines, China, and Korea in 2019",
"Heavy rainfall and flooding from Typhoon in Philippines, China, and Korea in 2019",
"Storm surges caused by typhoons in the Philippines, China, and Korea in 2019 resulting in coastal erosion and damage",
"People evacuating to safer areas during Typhoon in the Philippines, China, and Korea in 2019",
"Debris and wreckage scattered on roads and buildings after Typhoon in Philippines, China, and Korea in 2019"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EB%A7%81%EB%A7%81_(2019%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Lingling, it makes sense to extract events that are commonly associated with a powerful typhoon, such as strong winds, heavy rainfall, and storm surges, which could be visualized in a video. These events are likely to have occurred in the affected countries, including the Philippines, China, and Korea, and can be reasonably captioned in a video frame.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Storm surges causing coastal erosion and damage\n4. People evacuating to safer areas\n5. Debris and wreckage scattered on roads and buildings",
"en_article_excerpt": "Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Lingling_(2019)",
"event_category": "disasters",
"event_name": "typhoon_lingling",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about Typhoon Lingling, it makes sense to extract events that are typical of a powerful typhoon's impact, such as strong winds, heavy rainfall, and storm surges, which could be visualized in a video. These events are likely to occur in the affected countries, including the Philippines, China, and Korea, and can be reasonably captioned in a video frame.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Storm surges causing coastal erosion and damage\n4. People evacuating to safer areas or shelters\n5. Debris and destroyed buildings scattered in the typhoon's aftermath",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Lingling, also known in the Philippines as Typhoon Liwayway, was a powerful typhoon that hit the Philippines, China, and Korea in 2019.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are typically associated with the aftermath or impact of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Lingling. The events extracted should be specific to the regions affected by the typhoon, including the Philippines, China, and Korea.\n\nEVENTS:\n1. Buildings and houses destroyed by strong winds\n2. Flooding in coastal areas and low-lying regions\n3. People evacuating to higher ground or shelters\n4. Trees and power lines uprooted or knocked down\n5. Rescue teams providing aid to affected communities",
"video_URL": "http://youtube.com/watch?v=t1uezBvD9U8",
"video_description": "태풍 '링링'이 서해안을 따라 빠르게 북상하면서 충북 지역 피해도 속출하고 있습니다. 충북 지역 피해 현장을 카메라에 담았습니다.",
"video_id": "t1uezBvD9U8"
}
| 4
|
[
"The image depicts a large tree that has fallen onto a sidewalk in an urban area. The tree appears to have been uprooted and is blocking part of the street. A person wearing a yellow jacket is walking on the sidewalk near the fallen tree. In the background, there are buildings with visible signage and a digital sign displaying Korean text. The overall scene suggests that the tree fell as a result of a strong wind or storm, disrupting the urban environment.",
"The current frame shows a damaged blue wall with a large tear, revealing an underlying structure. In front of the wall, there is a device with yellow cylindrical objects and red wires, likely part of a fallen or malfunctioning sign. The scene suggests ongoing damage, possibly from the same storm or strong wind that caused the tree to fall in the previous frame. The blue wall is situated in an urban area with visible signage in Korean, indicating that this damage is part of the same incident.",
"The scene depicts a rainy day in an urban setting with a focus on a damaged netting structure, possibly part of a sports facility or a protective barrier. In the background, a red sign with Korean text is visible, indicating a commercial or public area. The netting appears to be dislodged or collapsed, likely due to strong winds or a storm. The presence of raindrops on the netting suggests recent or ongoing rainfall. The overall scene indicates the aftermath of a weather event, similar to the damage described in the previous frame.",
"The scene continues to depict a rainy day in an urban setting with the focus remaining on the damaged netting structure, likely part of a sports facility or a protective barrier. The netting remains dislodged or collapsed, indicating it has not been repaired or altered since the previous frame. Raindrops are still visible on the netting, reinforcing the ongoing rainfall. The red sign with Korean text in the background remains a focal point, suggesting the location is still within a commercial or public area. The overall scene continues to suggest the aftermath of a weather event, consistent with the damage seen in the previous frames."
] |
t1uezBvD9U8
|
A severe storm or strong wind event has caused significant damage in an urban area, likely in Korea given the presence of Korean signage. The storm uprooted a large tree, which fell onto a sidewalk, blocking part of the street. Additionally, it damaged a blue wall, revealing its underlying structure, and dislodged or collapsed a netting structure, possibly part of a sports facility or protective barrier. The damage occurred during rainy conditions, with raindrops visible on the netting in later frames, indicating ongoing rainfall. The scenes suggest that the incident disrupted the urban environment, causing destruction to both natural and man-made structures, and that the area is a commercial or public space given the presence of signage and possibly a sports facility.
|
Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.
|
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding in urban areas of the Korean Peninsula during Typhoon Soulik in 2012",
"Heavy rainfall causing flooding in urban areas of the Korean Peninsula during Typhoon Khanun in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Soulik in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Khanun in 2012"
] |
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities caused by Typhoon Soulik on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities due to Typhoon Khanun on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Soulik hit the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Khanun hit the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
[
"Strong winds from Typhoon Soulik blowing down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun blowing down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Soulik on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Khanun in the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Soulik on the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Khanun on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Soulik on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Khanun in the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EC%86%94%EB%A6%AD_(2018%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Soulik, it makes sense to extract events that are commonly associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and damage to infrastructure. These events are likely to be visualized in a video and are consistent with the characteristics of a powerful typhoon like Soulik.\n\nEVENTS:\n1. Strong winds blowing down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Debris flying through the air\n4. Roofs being torn off buildings\n5. Emergency responders evacuating people from flooded areas",
"en_article_excerpt": "Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Soulik_(2018)",
"event_category": "disasters",
"event_name": "typhoon_soulik",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to Typhoon Soulik, it makes sense to consider the typical consequences and scenes associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and resulting damage. Given the typhoon's impact on the Korean Peninsula, events related to severe weather, destruction, and potential evacuation or rescue efforts are likely to be relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Residents evacuating their homes and seeking shelter\n4. Emergency responders rescuing people from flooded areas\n5. Debris and damaged buildings scattered across the affected regions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact and aftermath of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Soulik. Given the severity of the typhoon, events that depict its powerful effects on the Korean Peninsula are most relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal areas and cities\n3. People evacuating their homes and seeking shelter\n4. Debris and destroyed buildings scattered across streets\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://twitter.com/i/status/1032900418777407489",
"video_description": "제주도 태풍 솔릭의 강풍의 무서움.. 오토바이도 날려버리는 태풍 솔릭.. ㄷㄷㄷ",
"video_id": "1032900418777407489"
}
| 4
|
[
"The image shows the interior view from inside a vehicle, looking out through a front windshield. The windshield wipers are in motion, suggesting that it is raining outside. The view outside is somewhat obscured by the rain and the windshield wipers' movement. Buildings and a street are visible in the background, indicating that the vehicle is likely parked or moving slowly in an urban area.",
"The current frame shows a continuation of the previous scene with a view from inside a vehicle, focusing on the front windshield. The windshield wipers are still active, indicating ongoing rain. The visibility is somewhat limited due to the rain and the wipers' motion. Outside, a person on a bicycle is seen moving across the street, suggesting that despite the weather, pedestrians and cyclists are navigating the urban environment. The background still includes buildings and a street, reinforcing the setting in a city during rainy conditions.",
"The current frame continues to show the view from inside the vehicle, with the windshield wipers still actively removing rain from the windshield. Visibility remains somewhat limited due to the rain and the wipers' motion. The scene outside now includes a red object, possibly a piece of clothing or a bag, attached to the rearview mirror, swinging with the vehicle's movement. This suggests the presence of a passenger or the driver's own belongings. The urban environment with buildings and streets remains consistent, indicating that the vehicle is still navigating through the city during the rain.",
"The scene continues to depict the interior view of the vehicle during a rainstorm, with the windshield wipers still in operation, maintaining visibility through the wet windshield. The red object attached to the rearview mirror remains visible, likely indicating the presence of a passenger or the driver's belongings. The exterior view shows the vehicle driving through a cityscape, characterized by buildings and streets, suggesting the vehicle is navigating through an urban area despite the rainy conditions."
] |
1032900418777407489
|
The video depicts a vehicle navigating through an urban area during a rainstorm. The scene is captured from the interior of the vehicle, looking out through the front windshield, with windshield wipers in constant motion to maintain visibility. As the vehicle moves slowly through the city, a cyclist is seen crossing the street, indicating that despite the rain, pedestrians and cyclists are present. The vehicle appears to have at least one occupant, suggested by a red object attached to the rearview mirror. Throughout the video, the consistent presence of buildings and streets in the background confirms the urban setting, showcasing daily life continuing amidst rainy weather conditions.
|
Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.
|
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding in urban areas of the Korean Peninsula during Typhoon Soulik in 2012",
"Heavy rainfall causing flooding in urban areas of the Korean Peninsula during Typhoon Khanun in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Soulik in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Khanun in 2012"
] |
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities caused by Typhoon Soulik on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities due to Typhoon Khanun on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Soulik hit the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Khanun hit the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
[
"Strong winds from Typhoon Soulik blowing down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun blowing down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Soulik on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Khanun in the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Soulik on the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Khanun on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Soulik on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Khanun in the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EC%86%94%EB%A6%AD_(2018%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Soulik, it makes sense to extract events that are commonly associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and damage to infrastructure. These events are likely to be visualized in a video and are consistent with the characteristics of a powerful typhoon like Soulik.\n\nEVENTS:\n1. Strong winds blowing down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Debris flying through the air\n4. Roofs being torn off buildings\n5. Emergency responders evacuating people from flooded areas",
"en_article_excerpt": "Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Soulik_(2018)",
"event_category": "disasters",
"event_name": "typhoon_soulik",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to Typhoon Soulik, it makes sense to consider the typical consequences and scenes associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and resulting damage. Given the typhoon's impact on the Korean Peninsula, events related to severe weather, destruction, and potential evacuation or rescue efforts are likely to be relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Residents evacuating their homes and seeking shelter\n4. Emergency responders rescuing people from flooded areas\n5. Debris and damaged buildings scattered across the affected regions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact and aftermath of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Soulik. Given the severity of the typhoon, events that depict its powerful effects on the Korean Peninsula are most relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal areas and cities\n3. People evacuating their homes and seeking shelter\n4. Debris and destroyed buildings scattered across streets\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://twitter.com/i/status/1032211760315420675",
"video_description": "실시간 제주도 서귀포 상황 태풍 솔릭...위험행..........조심하세요다들",
"video_id": "1032211760315420675"
}
| 4
|
[
"The video shows a scene from inside a vehicle, looking out through the windshield. Palm trees are swaying violently, indicating strong winds, and a fallen palm tree is blocking part of the road. There are other vehicles on the road, and the environment appears to be a coastal or tropical area.",
"The scene continues from inside the vehicle, showing a significant accumulation of palm fronds and debris on the road, which may have been caused by a fallen palm tree. The strong winds persist, as evidenced by the movement of the remaining trees and debris. The vehicle remains stationary, indicating that the road is obstructed, and the environment still appears to be a coastal or tropical area with palm trees.",
"The vehicle remains stationary, showing an obstructed road due to the accumulation of palm fronds and debris, likely from a fallen palm tree. The strong winds continue to move the remaining trees and debris, indicating the ongoing impact of a storm or tropical weather.",
"The vehicle remains stationary, showing an obstructed road due to the accumulation of palm fronds and debris, likely from a fallen palm tree. The strong winds continue to move the remaining trees and debris, indicating the ongoing impact of a storm or tropical weather."
] |
1032211760315420675
|
The video depicts a scene from inside a vehicle stuck on a road in a coastal or tropical area during a storm. The road is obstructed by debris, including palm fronds and a fallen palm tree, caused by strong winds. The vehicle remains stationary throughout, unable to proceed due to the blockage, as the storm continues to affect the environment with persisting strong winds and swaying trees.
|
Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.
|
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding in urban areas of the Korean Peninsula during Typhoon Soulik in 2012",
"Heavy rainfall causing flooding in urban areas of the Korean Peninsula during Typhoon Khanun in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Soulik in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Khanun in 2012"
] |
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities caused by Typhoon Soulik on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities due to Typhoon Khanun on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Soulik hit the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Khanun hit the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
[
"Strong winds from Typhoon Soulik blowing down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun blowing down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Soulik on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Khanun in the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Soulik on the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Khanun on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Soulik on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Khanun in the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EC%86%94%EB%A6%AD_(2018%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Soulik, it makes sense to extract events that are commonly associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and damage to infrastructure. These events are likely to be visualized in a video and are consistent with the characteristics of a powerful typhoon like Soulik.\n\nEVENTS:\n1. Strong winds blowing down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Debris flying through the air\n4. Roofs being torn off buildings\n5. Emergency responders evacuating people from flooded areas",
"en_article_excerpt": "Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Soulik_(2018)",
"event_category": "disasters",
"event_name": "typhoon_soulik",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to Typhoon Soulik, it makes sense to consider the typical consequences and scenes associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and resulting damage. Given the typhoon's impact on the Korean Peninsula, events related to severe weather, destruction, and potential evacuation or rescue efforts are likely to be relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Residents evacuating their homes and seeking shelter\n4. Emergency responders rescuing people from flooded areas\n5. Debris and damaged buildings scattered across the affected regions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact and aftermath of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Soulik. Given the severity of the typhoon, events that depict its powerful effects on the Korean Peninsula are most relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal areas and cities\n3. People evacuating their homes and seeking shelter\n4. Debris and destroyed buildings scattered across streets\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://twitter.com/i/status/1032540315515682816",
"video_description": "#태풍 #솔릭 예상 경로! 동쪽으로 방향 틀어",
"video_id": "1032540315515682816"
}
| 4
|
[
"The image shows a weather map from windy.com indicating temperature variations across East Asia, including parts of North Korea, South Korea, and Japan. The map uses color gradients to represent temperature, with cooler temperatures in blue and warmer temperatures in red and yellow. Various city names are labeled with their corresponding temperatures, such as Seoul (26°C), Tokyo (27°C), and Beijing (24°C). The map highlights temperature variations across the region.",
"The weather map continues to display temperature variations across East Asia, with North Korea and South Korea prominently featured. The color gradients persist, showing cooler temperatures in blue and warmer temperatures in red and yellow. City names with their respective temperatures remain visible, such as Seoul (26°C), Tokyo (27°C), and Beijing (24°C), reinforcing the focus on temperature differences across the region.",
"The weather map continues to display temperature variations across East Asia, with North Korea and South Korea prominently featured. The color gradients persist, showing cooler temperatures in blue and warmer temperatures in red and yellow. City names with their respective temperatures remain visible, such as Seoul (22°C), Tokyo (26°C), and Beijing (25°C), reinforcing the focus on temperature differences across the region.",
"The weather map shows temperature variations across East Asia, with North Korea and South Korea prominently featured. The color gradients persist, showing cooler temperatures in blue and warmer temperatures in red and yellow. City names with their respective temperatures remain visible, such as Seoul (28°C), Tokyo (26°C), and Beijing (24°C), reinforcing the focus on temperature differences across the region."
] |
1032540315515682816
|
The video depicts a weather map of East Asia, focusing on temperature variations across the region, particularly in North Korea, South Korea, and Japan. The map displays city names along with their respective temperatures, which fluctuate across frames, such as Seoul, Tokyo, and Beijing. Throughout the video, the map uses color gradients to represent temperature differences, with cooler temperatures in blue and warmer temperatures in red and yellow. The overall context suggests a weather forecast or temperature update for the East Asian region.
|
Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.
|
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding in urban areas of the Korean Peninsula during Typhoon Soulik in 2012",
"Heavy rainfall causing flooding in urban areas of the Korean Peninsula during Typhoon Khanun in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Soulik in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Khanun in 2012"
] |
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities caused by Typhoon Soulik on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities due to Typhoon Khanun on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Soulik hit the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Khanun hit the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
[
"Strong winds from Typhoon Soulik blowing down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun blowing down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Soulik on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Khanun in the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Soulik on the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Khanun on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Soulik on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Khanun in the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EC%86%94%EB%A6%AD_(2018%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Soulik, it makes sense to extract events that are commonly associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and damage to infrastructure. These events are likely to be visualized in a video and are consistent with the characteristics of a powerful typhoon like Soulik.\n\nEVENTS:\n1. Strong winds blowing down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Debris flying through the air\n4. Roofs being torn off buildings\n5. Emergency responders evacuating people from flooded areas",
"en_article_excerpt": "Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Soulik_(2018)",
"event_category": "disasters",
"event_name": "typhoon_soulik",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to Typhoon Soulik, it makes sense to consider the typical consequences and scenes associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and resulting damage. Given the typhoon's impact on the Korean Peninsula, events related to severe weather, destruction, and potential evacuation or rescue efforts are likely to be relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Residents evacuating their homes and seeking shelter\n4. Emergency responders rescuing people from flooded areas\n5. Debris and damaged buildings scattered across the affected regions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact and aftermath of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Soulik. Given the severity of the typhoon, events that depict its powerful effects on the Korean Peninsula are most relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal areas and cities\n3. People evacuating their homes and seeking shelter\n4. Debris and destroyed buildings scattered across streets\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://twitter.com/i/status/1032471495958618112",
"video_description": "어제 저녁부터 비바람이 거세지다 잠잠하다 하더니, 지금은 세차게 몰아치는 중입니다 ㅠㅠ #태풍솔릭 #제주도 #제주시",
"video_id": "1032471495958618112"
}
| 4
|
[
"The current frame shows a large tree with dense green foliage. The setting appears to be outdoors, likely in a park or garden, with the ground visible below the tree, suggesting a natural environment. The sky is overcast, indicating a cloudy day.",
"The current frame continues to depict a large tree with dense green foliage. The setting remains consistent with the previous frame, indicating an outdoor location, likely a park or garden. The ground below the tree is still visible, reinforcing the natural environment. The sky remains overcast, suggesting that the cloudy day from the previous frame persists, maintaining the overall ambiance of the scene.",
"The current frame shows a person standing near the base of a large tree with dense green foliage, indicating that they might be observing or tending to the tree. The setting continues to depict an outdoor location, likely a park or garden. The ground below the tree is still visible, reinforcing the natural environment. The sky remains overcast, suggesting that the cloudy day from the previous frames persists, maintaining the overall ambiance of the scene.",
"The focus remains on the large tree with dense green foliage. The person is no longer visible, indicating they may have moved out of the frame or are obscured by the tree. The setting continues to depict an outdoor location, likely a park or garden. The ground below the tree is still visible, reinforcing the natural environment. The sky remains overcast, suggesting that the cloudy day persists, maintaining the overall ambiance of the scene."
] |
1032471495958618112
|
The video depicts a scene in a park or garden on a cloudy day, focusing on a large tree with dense green foliage. A person is introduced, standing near the base of the tree, possibly observing or tending to it, before moving out of view. The setting remains consistent throughout, with the tree and natural environment as the central elements, under an overcast sky.
|
Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.
|
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding in urban areas of the Korean Peninsula during Typhoon Soulik in 2012",
"Heavy rainfall causing flooding in urban areas of the Korean Peninsula during Typhoon Khanun in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Soulik in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Khanun in 2012"
] |
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities caused by Typhoon Soulik on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities due to Typhoon Khanun on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Soulik hit the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Khanun hit the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
[
"Strong winds from Typhoon Soulik blowing down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun blowing down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Soulik on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Khanun in the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Soulik on the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Khanun on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Soulik on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Khanun in the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EC%86%94%EB%A6%AD_(2018%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Soulik, it makes sense to extract events that are commonly associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and damage to infrastructure. These events are likely to be visualized in a video and are consistent with the characteristics of a powerful typhoon like Soulik.\n\nEVENTS:\n1. Strong winds blowing down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Debris flying through the air\n4. Roofs being torn off buildings\n5. Emergency responders evacuating people from flooded areas",
"en_article_excerpt": "Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Soulik_(2018)",
"event_category": "disasters",
"event_name": "typhoon_soulik",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to Typhoon Soulik, it makes sense to consider the typical consequences and scenes associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and resulting damage. Given the typhoon's impact on the Korean Peninsula, events related to severe weather, destruction, and potential evacuation or rescue efforts are likely to be relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Residents evacuating their homes and seeking shelter\n4. Emergency responders rescuing people from flooded areas\n5. Debris and damaged buildings scattered across the affected regions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact and aftermath of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Soulik. Given the severity of the typhoon, events that depict its powerful effects on the Korean Peninsula are most relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal areas and cities\n3. People evacuating their homes and seeking shelter\n4. Debris and destroyed buildings scattered across streets\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://twitter.com/i/status/1032234901897015298",
"video_description": "일본 가고시마에 상륙한 태풍 솔릭 위력ㅠㅠㅠ 8월19일자",
"video_id": "1032234901897015298"
}
| 4
|
[
"The image depicts a scene of a storm with heavy rain and strong winds in a residential area. A small yellow car is parked on the street, and a person is standing beside it, seemingly holding onto a bicycle. The street is wet, and the visibility is reduced due to the rain. The text on the image indicates that this is footage from a place called Nanki Awaji, captured before 6 p.m.",
"The scene continues to show the storm's impact, with heavy rain and strong winds persisting in the residential area. The small yellow car remains parked on the street, and the person is still beside it, holding onto the bicycle. The visibility remains low due to the rain, and the text on the image suggests that this footage is from a location called Amano Shima, captured at 6:16 p.m.",
"The storm continues to affect the residential area with persistent heavy rain and strong winds. The small yellow car remains parked on the street, and the person is still holding onto the bicycle. Visibility remains low due to the rain. Text on the image indicates that the footage is from Amano Shima, captured at 7:00 p.m., and mentions a \"very strong typhoon,\" with warnings about the storm's impact.",
"The storm persists in the residential area with continuous heavy rain and strong winds. The small yellow car is still parked on the street, and the person is still holding onto the bicycle. Visibility remains low due to the rain. Text on the image indicates that the footage is from Amano Shima, captured at 7:00 p.m., and mentions a \"very strong typhoon,\" with warnings about the storm's impact. A map in the corner shows the typhoon's path and intensity."
] |
1032234901897015298
|
A severe storm, identified as a very strong typhoon, hits a residential area, causing heavy rain and strong winds. The storm affects locations including Nanki Awaji and Amano Shima, with footage captured between before 6 p.m. and 7 p.m. A person is seen beside a parked small yellow car, holding onto a bicycle, struggling with the low visibility and harsh weather conditions. The typhoon's path and intensity are tracked on a map, with warnings issued about its impact, indicating a significant and potentially dangerous weather event.
|
Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.
|
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding in urban areas of the Korean Peninsula during Typhoon Soulik in 2012",
"Heavy rainfall causing flooding in urban areas of the Korean Peninsula during Typhoon Khanun in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Soulik in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Khanun in 2012"
] |
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities caused by Typhoon Soulik on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities due to Typhoon Khanun on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Soulik hit the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Khanun hit the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
[
"Strong winds from Typhoon Soulik blowing down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun blowing down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Soulik on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Khanun in the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Soulik on the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Khanun on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Soulik on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Khanun in the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EC%86%94%EB%A6%AD_(2018%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Soulik, it makes sense to extract events that are commonly associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and damage to infrastructure. These events are likely to be visualized in a video and are consistent with the characteristics of a powerful typhoon like Soulik.\n\nEVENTS:\n1. Strong winds blowing down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Debris flying through the air\n4. Roofs being torn off buildings\n5. Emergency responders evacuating people from flooded areas",
"en_article_excerpt": "Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Soulik_(2018)",
"event_category": "disasters",
"event_name": "typhoon_soulik",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to Typhoon Soulik, it makes sense to consider the typical consequences and scenes associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and resulting damage. Given the typhoon's impact on the Korean Peninsula, events related to severe weather, destruction, and potential evacuation or rescue efforts are likely to be relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Residents evacuating their homes and seeking shelter\n4. Emergency responders rescuing people from flooded areas\n5. Debris and damaged buildings scattered across the affected regions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact and aftermath of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Soulik. Given the severity of the typhoon, events that depict its powerful effects on the Korean Peninsula are most relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal areas and cities\n3. People evacuating their homes and seeking shelter\n4. Debris and destroyed buildings scattered across streets\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://youtube.com/watch?v=_V6HiUPBShQ",
"video_description": "태풍 피해 전해드렸는데요, 이렇게 제주와 일부 내륙에 피해가 집중된 건 예보와 달리 솔릭의 위력이 급격히 떨어졌기 때문입니다. 시속 100km를 넘나드는 강풍과 비를 동반해 모두를 긴장시켰던 태풍이 왜 이렇게 빠르게 약해졌는지, 엄민재 기자가 전해드립니다. 뉴스 원문 기사 보기 https://news.sbs.co.kr/news/endPage.do?news_id=N1004904932 [태풍 '솔릭' 북상] 기사 더보기 https://news.sbs.co.kr/news/newsHotIssueList.do?tagId=10000046975 #SBS뉴스 #SBSNEWS #SBS_NEWS #에스비에스뉴스 #태풍_'솔릭'_북상 #8뉴스 ▶SBS 뉴스 채널 구독하기 : https://n.sbs.co.kr/youtube ▶SBS 뉴스 라이브 : https://n.sbs.co.kr/youtubeLive , https://n.sbs.co.kr/live ▶대한민국 뉴스리더 SBS◀ 홈페이지: https://n.sbs.co.kr 페이스북: https://www.facebook.com/sbs8news 트위터: https://www.twitter.com/sbs8news",
"video_id": "_V6HiUPBShQ"
}
| 4
|
[
"The image shows a satellite view of two large tropical cyclones over the ocean, near the East Asian continent. The cyclones have distinct spiral cloud patterns with visible eye centers. The outlines of several countries are marked in yellow, including parts of China, Japan, and the Korean Peninsula. The image is from a weather news broadcast, as indicated by the \"SBS NEWS\" logo in the bottom left corner.",
"The current frame shows a group of palm trees heavily swaying in the wind, indicating strong wind conditions, likely due to a tropical cyclone. The background is overcast, suggesting stormy weather. The text \"급격히 힘 빠진 솔릭, 왜?\" (Suddenly weakened Solik, why?) suggests a discussion on the weakening of a tropical cyclone named Solik. The \"SBS NEWS\" logo is visible, maintaining the context of a weather news broadcast.",
"The current frame shows a satellite image of a large body of water, likely the ocean, with a clearly defined tropical cyclone positioned near the eastern coast of a continent. The cyclone is depicted with a distinct spiral pattern, typical of such weather systems. The surrounding region includes landmasses with varying shades of green, indicating different types of terrain and vegetation. The text \"급격히 힘 빠진 솔릭, 왜?\" (Suddenly weakened Solik, why?) is visible, indicating that the video is discussing the rapid weakening of the cyclone named Solik. The \"SBS NEWS\" logo is present, reinforcing the context of a weather news broadcast. The map is labeled \"19호 태풍 솔릭\" (Typhoon Solik No. 19), identifying the specific cyclone being discussed.",
"The current frame continues to show a satellite image of the ocean with the tropical cyclone, still positioned near the eastern coast of the continent. The cyclone's spiral pattern is less pronounced than in the previous frame, suggesting a further weakening. The surrounding landmasses remain visible with their varying shades of green. The text \"급격히 힘 빠진 솔릭, 왜?\" (Suddenly weakened Solik, why?) is still present, maintaining the focus on the cyclone's rapid weakening. The \"SBS NEWS\" logo continues to be displayed, reinforcing the ongoing weather news broadcast. The map label \"19호 태풍 솔릭\" (Typhoon Solik No. 19) remains, identifying the cyclone being discussed."
] |
_V6HiUPBShQ
|
A weather news broadcast discusses the rapid weakening of Typhoon Solik, a tropical cyclone located near the eastern coast of East Asia. The video shows satellite images of the cyclone over the ocean, with its distinct spiral cloud pattern and visible eye center, as well as its impact on land, including strong winds causing palm trees to sway heavily. The broadcast explores the unexpected weakening of the cyclone, labeled as "19호 태풍 솔릭" (Typhoon Solik No. 19), and investigates the reasons behind this sudden change, as indicated by the text "급격히 힘 빠진 솔릭, 왜?" (Suddenly weakened Solik, why?).
|
Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.
|
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding in urban areas of the Korean Peninsula during Typhoon Soulik in 2012",
"Heavy rainfall causing flooding in urban areas of the Korean Peninsula during Typhoon Khanun in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Soulik in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Khanun in 2012"
] |
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities caused by Typhoon Soulik on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities due to Typhoon Khanun on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Soulik hit the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Khanun hit the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
[
"Strong winds from Typhoon Soulik blowing down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun blowing down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Soulik on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Khanun in the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Soulik on the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Khanun on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Soulik on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Khanun in the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EC%86%94%EB%A6%AD_(2018%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Soulik, it makes sense to extract events that are commonly associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and damage to infrastructure. These events are likely to be visualized in a video and are consistent with the characteristics of a powerful typhoon like Soulik.\n\nEVENTS:\n1. Strong winds blowing down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Debris flying through the air\n4. Roofs being torn off buildings\n5. Emergency responders evacuating people from flooded areas",
"en_article_excerpt": "Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Soulik_(2018)",
"event_category": "disasters",
"event_name": "typhoon_soulik",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to Typhoon Soulik, it makes sense to consider the typical consequences and scenes associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and resulting damage. Given the typhoon's impact on the Korean Peninsula, events related to severe weather, destruction, and potential evacuation or rescue efforts are likely to be relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Residents evacuating their homes and seeking shelter\n4. Emergency responders rescuing people from flooded areas\n5. Debris and damaged buildings scattered across the affected regions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact and aftermath of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Soulik. Given the severity of the typhoon, events that depict its powerful effects on the Korean Peninsula are most relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal areas and cities\n3. People evacuating their homes and seeking shelter\n4. Debris and destroyed buildings scattered across streets\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://youtube.com/watch?v=Aiq8I97bscw",
"video_description": "제19호 태풍 '솔릭'이 일본 남부 해상에서 한반도를 향해 북진하고 있습니다. '솔릭'은 강한 중형급 태풍으로 중심기압은 960hPa, 강풍 반경은 340㎞이라고 하는데요. 만일 태풍이 우리나라를 관통하면 2012년 9월 '산바' 이후 약 6년 만으로 큰 피해가 우려됩니다. 자세한 내용 영상으로 보시죠. ◆ 연합뉴스 홈페이지→ http://www.yna.co.kr/ ◆ 이 시각 많이 본 기사 → https://goo.gl/VQTsSZ ◆ 오늘의 핫뉴스 → https://goo.gl/WyGXpG ◆ 현장영상 → https://goo.gl/5aZcx8 ◆ 카드뉴스 →https://goo.gl/QKfDTH ◆연합뉴스 공식 SNS◆ ◇페이스북→ https://www.facebook.com/yonhap/ ▣ 연합뉴스 유튜브 채널 구독 : https://goo.gl/pL7TmT",
"video_id": "Aiq8I97bscw"
}
| 4
|
[
"A stormy scene shows a road engulfed in heavy fog with low visibility. Vehicles are parked on the side of the road, and a large barrel lies on the ground. In the background, a building is partially visible through the mist. The text in the image indicates that the storm has strong winds, with a central pressure of 960hPa and a wind range of 340km.",
"The current frame depicts a powerful storm at sea with turbulent waves crashing violently. The text in the image suggests it is a depiction of a tornado with a wind speed of 110 meters per second, likely emphasizing the intense and dangerous conditions created by the storm.",
"The current frame shows a view of a flooded street with multiple cars driving cautiously through the water. The scene depicts challenging driving conditions due to heavy rainfall, with vehicles turning on their headlights to navigate through the foggy and wet environment. The text in the image indicates that there is a possibility of a dangerous situation occurring, advising people to be cautious, likely highlighting the potential hazards associated with flooding and adverse weather conditions.",
"The current frame displays a logo for Yonhap News Agency against a blurred background. The logo is prominently centered, suggesting a transition or an introduction to a news segment. The background appears to be a continuation of the previous scene, indicating that the news segment may still be discussing the flooding conditions depicted earlier. The logo's presence typically signifies a news broadcast or an informational segment, likely providing updates or details on the ongoing situation."
] |
Aiq8I97bscw
|
A severe storm is depicted, bringing heavy fog, strong winds, and flooding to a coastal area. The storm's intense conditions, including a tornado with high wind speeds, pose significant dangers. Vehicles are shown navigating through flooded streets with caution, highlighting the challenging driving conditions. The scene transitions to a news segment introduced by the Yonhap News Agency logo, suggesting that the video is a news report providing updates and information on the storm and its impact.
|
Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.
|
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding in urban areas of the Korean Peninsula during Typhoon Soulik in 2012",
"Heavy rainfall causing flooding in urban areas of the Korean Peninsula during Typhoon Khanun in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Soulik in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Khanun in 2012"
] |
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities caused by Typhoon Soulik on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities due to Typhoon Khanun on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Soulik hit the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Khanun hit the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
[
"Strong winds from Typhoon Soulik blowing down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun blowing down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Soulik on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Khanun in the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Soulik on the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Khanun on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Soulik on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Khanun in the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EC%86%94%EB%A6%AD_(2018%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Soulik, it makes sense to extract events that are commonly associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and damage to infrastructure. These events are likely to be visualized in a video and are consistent with the characteristics of a powerful typhoon like Soulik.\n\nEVENTS:\n1. Strong winds blowing down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Debris flying through the air\n4. Roofs being torn off buildings\n5. Emergency responders evacuating people from flooded areas",
"en_article_excerpt": "Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Soulik_(2018)",
"event_category": "disasters",
"event_name": "typhoon_soulik",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to Typhoon Soulik, it makes sense to consider the typical consequences and scenes associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and resulting damage. Given the typhoon's impact on the Korean Peninsula, events related to severe weather, destruction, and potential evacuation or rescue efforts are likely to be relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Residents evacuating their homes and seeking shelter\n4. Emergency responders rescuing people from flooded areas\n5. Debris and damaged buildings scattered across the affected regions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact and aftermath of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Soulik. Given the severity of the typhoon, events that depict its powerful effects on the Korean Peninsula are most relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal areas and cities\n3. People evacuating their homes and seeking shelter\n4. Debris and destroyed buildings scattered across streets\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://youtube.com/watch?v=25LVZnXdx0A",
"video_description": "중부지방을 중심으로 연일 매섭게 비가 쏟아지고 있다. 긴급상황에 대비해야 할 소방서까지 침수된 모습이 SNS를 통해 전파되면서 불안이 가중되고 있다. 비도 이 정도면 재해 수준. 아무 일도 없다는 듯 태풍 솔릭은 지나갔지만, 오히려 비의 공포는 더해져만 가는데... 제작 : 박태호PD(ptho@ytnplus.co.kr) ▶ 기사 원문 : http://www.ytn.co.kr/_ln/0103_201808292100061801 ▶ 제보 안내 : http://goo.gl/gEvsAL, 모바일앱, 8585@ytn.co.kr, #2424 ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ [ 한국 뉴스 채널 와이티엔 / Korea News Channel YTN ]",
"video_id": "25LVZnXdx0A"
}
| 4
|
[
"The image shows a flooded indoor area, possibly a store or a warehouse, with water covering the floor. Clothing racks and shelves are visible above the waterline. The text in Korean reads: \"위터파크를 연상시키는 이 곳은...?\" which translates to \"This place reminds me of Water Park...?\"",
"The current frame shows a flooded indoor area, likely a kitchen or living space, with water covering the floor. The water level appears to be significantly high, submerging part of the kitchen appliances and furniture. A person is visible standing in the doorway, observing the flooded area. The text in Korean reads: \"침수 피해로 강제 물놀이 행\" which translates to \"Forced water play due to flood damage.\" The scene continues to depict the aftermath of a flood, emphasizing the extent of water damage and the unexpected turn of events for the occupants.",
"The video shifts to show a turbulent river with muddy brown water, indicating severe flooding. The river is flowing rapidly, with visible white foam and debris, suggesting heavy rainfall and possibly a nearby storm. The background includes a barrier with a fence on top, separating the river from a higher ground area with some greenery and trees. The text 'YTN' is visible in the top right corner, likely indicating the broadcasting network.",
"The current frame displays a black background with white text in Korean, which reads \"물 조심,\" translating to \"Be careful with water.\" This text serves as a warning, likely cautioning viewers about the dangers of the flooding depicted in the previous frames, emphasizing the importance of safety in the face of severe weather conditions."
] |
25LVZnXdx0A
|
The video depicts the aftermath of a severe flood, showing indoor areas, such as a store or warehouse and a kitchen, completely inundated with water, with furniture and appliances partially submerged. A person is seen observing the damage from a doorway. The flood's origin is then shown as a turbulent, muddy river with rapid flow and debris, indicating heavy rainfall and a possible storm. The video concludes with a cautionary message, warning viewers to be careful with water, emphasizing the dangers of flooding and the importance of safety during severe weather conditions.
|
Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.
|
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding in urban areas of the Korean Peninsula during Typhoon Soulik in 2012",
"Heavy rainfall causing flooding in urban areas of the Korean Peninsula during Typhoon Khanun in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"Residents evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Soulik in 2012",
"Debris and damaged buildings scattered across the Korean Peninsula after Typhoon Khanun in 2012"
] |
[
"Strong winds from Typhoon Soulik knocking down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun knocking down trees and power lines on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities caused by Typhoon Soulik on the Korean Peninsula in 2012",
"Flooding in coastal areas and cities due to Typhoon Khanun on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Soulik on the Korean Peninsula in 2012",
"People evacuating their homes and seeking shelter during Typhoon Khanun on the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Soulik hit the Korean Peninsula in 2012",
"Debris and destroyed buildings scattered across streets after Typhoon Khanun hit the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders rescuing people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
[
"Strong winds from Typhoon Soulik blowing down trees and power lines on the Korean Peninsula in 2012",
"Strong winds from Typhoon Khanun blowing down trees and power lines on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Soulik on the Korean Peninsula in 2012",
"Heavy rainfall and flooding from Typhoon Khanun in the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Soulik on the Korean Peninsula in 2012",
"Debris flying through the air during Typhoon Khanun on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Soulik on the Korean Peninsula in 2012",
"Roofs being torn off buildings during Typhoon Khanun in the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Soulik on the Korean Peninsula in 2012",
"Emergency responders evacuating people from flooded areas during Typhoon Khanun on the Korean Peninsula in 2012"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%83%9C%ED%92%8D_%EC%86%94%EB%A6%AD_(2018%EB%85%84)",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about Typhoon Soulik, it makes sense to extract events that are commonly associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and damage to infrastructure. These events are likely to be visualized in a video and are consistent with the characteristics of a powerful typhoon like Soulik.\n\nEVENTS:\n1. Strong winds blowing down trees and power lines\n2. Heavy rainfall flooding streets and homes\n3. Debris flying through the air\n4. Roofs being torn off buildings\n5. Emergency responders evacuating people from flooded areas",
"en_article_excerpt": "Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.",
"en_article_url": "https://en.wikipedia.org/wiki/Typhoon_Soulik_(2018)",
"event_category": "disasters",
"event_name": "typhoon_soulik",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to Typhoon Soulik, it makes sense to consider the typical consequences and scenes associated with a large and deadly typhoon, such as strong winds, heavy rainfall, and resulting damage. Given the typhoon's impact on the Korean Peninsula, events related to severe weather, destruction, and potential evacuation or rescue efforts are likely to be relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Heavy rainfall causing flooding in urban areas\n3. Residents evacuating their homes and seeking shelter\n4. Emergency responders rescuing people from flooded areas\n5. Debris and damaged buildings scattered across the affected regions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Typhoon Soulik was an unusually large, and the deadliest typhoon to strike the Korean Peninsula since Khanun in 2012.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact and aftermath of a typhoon, such as destruction, flooding, and evacuation efforts, as these events are likely to be visualized in a video about Typhoon Soulik. Given the severity of the typhoon, events that depict its powerful effects on the Korean Peninsula are most relevant.\n\nEVENTS:\n1. Strong winds knocking down trees and power lines\n2. Flooding in coastal areas and cities\n3. People evacuating their homes and seeking shelter\n4. Debris and destroyed buildings scattered across streets\n5. Emergency responders rescuing people from flooded areas",
"video_URL": "http://youtube.com/watch?v=BRxOTPDR0g4",
"video_description": "건물 공사 구조물이 무너지고 교회 첨탑과 호텔 지붕도 강풍에 날려 추락했습니다. 아름드리 가로수는 뿌리째 뽑혔습니다. 지난 2010년 태풍 곤파스가 몰고 온 강풍의 위력입니다. 전남 홍도에 초속 42.9m의 폭풍이 불었고 태안 38.1m, 군산 33.6m, 서울도 21.6m의 강풍이 기록됐습니다. 바람 태풍 곤파스로 6명이 숨지고 11명이 부상했으며, 1,673억 원의 재산 피해가 발생했습니다. 북상하는 19호 태풍 '솔릭'은 8년 전 '곤파스'와 비슷한 경로로 움직이고 있습니다. 내일 새벽, 제주도 서쪽을 거쳐 서해안을 따라 북상한 뒤 태안반도에 상륙해 수도권을 통과할 것으로 예상됩니다. 한반도 주변 수온이 28도 정도로 무척 높아, 곤파스보다 강한 세력을 유지한 채 북상할 가능성이 큽니다. 기상청은 태풍 '솔릭'으로 인해 목요일과 금요일 사이 전국에 최고 400mm의 폭우가 쏟아지고 초속 40m가 넘는 강풍이 불 것으로 전망했습니다. 취재기자 : 정혜윤 영상편집 : 연진영 자막뉴스 제작 : 육지혜 에디터 ▶ 기사 원문 : http://www.ytn.co.kr/_ln/0108_201808220940500446 ▶ 제보 안내 : http://goo.gl/gEvsAL, 모바일앱, 8585@ytn.co.kr, #2424 ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ [ 한국 뉴스 채널 와이티엔 / Korea News Channel YTN ]",
"video_id": "BRxOTPDR0g4"
}
| 4
|
[
"The image shows a weather news report from YTN with a focus on strong winds in various regions. It mentions record-breaking gusts, including 42.9 m/s in Hongdo, 38.1 m/s in Taean, 33.6 m/s in Gunsan, and 21.6 m/s in Seoul. There is a map of Korea with wind speed indicators marked. The background appears to show a building affected by the wind, with parts of the structure potentially damaged.",
"The current frame depicts a stormy seascape with rough waves crashing against a rocky shore. The camera is positioned behind a window with raindrops on it, suggesting heavy rainfall. The text in the frame provides information about Typhoon 'Solik' moving north, having traveled 8 years and now showing signs of weakening. The overall scene indicates a coastal area experiencing severe weather conditions due to the typhoon.",
"The current frame shows a ship navigating through rough seas under foggy conditions. The text on the screen suggests the weather is severe, with the wind speed at 28 meters per second and visibility significantly reduced, making navigation extremely challenging. The ship appears to be a cargo vessel, possibly involved in transporting goods or materials despite the harsh weather conditions. The overcast sky and the heavy fog indicate that the storm is still affecting the area, possibly as part of the aftermath of Typhoon 'Solik'.",
"The ship continues to navigate through the rough seas under the persistently foggy conditions. The challenging weather conditions, marked by high wind speeds and low visibility, make the journey treacherous. The cargo vessel remains on course, battling against the elements, likely still engaged in its mission to transport goods or materials. The overcast sky and dense fog persist, indicating that the storm's influence continues to impact the area, potentially part of the ongoing aftermath of Typhoon 'Solik'."
] |
BRxOTPDR0g4
|
A weather report from YTN highlights strong winds in various regions of Korea due to Typhoon 'Solik', with record-breaking gusts recorded in several areas. The typhoon brings severe weather conditions, including heavy rainfall and rough seas, to coastal areas, causing potential damage to structures. A cargo ship is shown navigating through the treacherous conditions, with high wind speeds and low visibility making the journey extremely challenging. Despite the harsh weather, the ship persists on its course, likely transporting goods or materials. The video depicts the intense impact of Typhoon 'Solik' on Korea, showcasing the storm's powerful winds, rough seas, and the efforts of a cargo vessel to operate amidst the severe conditions.
|
The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.
|
[
"Strong winds blowing sand and dust across the Eastern Gobi desert steppe during the sandstorm on March 14, 2021",
"Reduced visibility on roads and highways due to heavy sand and dust in the air during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"People wearing masks and protective gear to shield themselves from the massive sandstorm that hit the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and other parts of East Asia on 14 March 2021",
"Sand and dust accumulating on buildings, streets, and vehicles during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"Flights and daily activities being disrupted due to a sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, and Korean Peninsula on March 14, 2021"
] |
[
"People wearing masks walking in low visibility conditions during a sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on 14 March 2021",
"Cars driving with headlights on during the day due to reduced visibility caused by the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on March 14, 2021",
"Sandstorm reducing visibility in urban areas of the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"Flights grounded or delayed due to sandstorm in East Asia on March 14, 2021",
"Residents sealing windows and doors to prevent sand entry during the 14 March 2021 sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia"
] |
[
"Sandstorm reducing visibility on roads and highways in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"People wearing masks to protect themselves from the sandstorm in East Asia on March 14, 2021",
"Trees and buildings covered in dust and sand during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on 14 March 2021",
"Cars and buses driving slowly due to low visibility during a sandstorm in the Eastern Gobi desert or East Asia on March 14 2021",
"Residents cleaning up sand and dust from streets and homes after the 14 March 2021 sandstorm in East Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%99%A9%EC%82%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects and progression of the sandstorm, such as reduced visibility, dust in the air, and people protecting themselves from the storm. These events are likely to be visualized in a video associated with the query, given the geographical scope and impact of the sandstorm.\n\nEVENTS:\n1. Dust and sand reducing visibility on roads and highways\n2. People wearing masks to protect themselves from the sandstorm\n3. Trees and buildings covered in dust and sand\n4. Cars and buses driving slowly due to low visibility\n5. Residents cleaning up sand and dust from streets and homes",
"en_article_excerpt": "The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.",
"en_article_url": "https://en.wikipedia.org/wiki/2021_East_Asia_sandstorm",
"event_category": "disasters",
"event_name": "2021_sandstorm",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the 2021 East Asia sandstorm, it makes sense to consider the typical visual consequences of a large-scale sandstorm, such as reduced visibility, strong winds, and massive amounts of dust and sand particles in the air. Events that could be visualized in a video would likely involve the storm's impact on the environment and daily life, rather than abstract or behind-the-scenes occurrences.\n\nEVENTS:\n1. Strong winds blowing sand and dust across the Eastern Gobi desert steppe\n2. Reduced visibility on roads and highways due to heavy sand and dust in the air\n3. People wearing masks and protective gear to shield themselves from the storm\n4. Sand and dust accumulating on buildings, streets, and vehicles\n5. Flights and daily activities being disrupted due to poor weather conditions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2021 East Asia sandstorm, it makes sense to extract events that are directly related to the effects and progression of the sandstorm, such as reduced visibility, transportation disruptions, and daily life impacts. These events should be concrete and visual, allowing them to be reasonably captioned in a video frame associated with the query.\n\nEVENTS:\n1. People wearing masks walking in low visibility conditions\n2. Cars driving with headlights on during the day due to reduced visibility\n3. Sandstorm reducing visibility in urban areas\n4. Flights being grounded or delayed at airports\n5. Residents sealing windows and doors to prevent sand entry",
"video_URL": "http://youtube.com/watch?v=Yddr-IKR_KA",
"video_description": "중국 둔황에 초대형 모래폭풍이 불어닥치면서눈 깜짝할 사이에 도시 전체가 암흑으로 뒤덮었습니다. [Ch.19] 사실을 보고 진실을 말합니다. 👍🏻 공식 홈페이지 http://news.tvchosun.com/ 👍🏻 공식 페이스북 https://www.facebook.com/tvchosunnews/ 👍🏻 공식 트위터 https://twitter.com/TVChosunNews * 뉴스제보 : 이메일(tvchosun@chosun.com), 카카오톡(tv조선제보), 전화(1661-0190)",
"video_id": "Yddr-IKR_KA"
}
| 4
|
[
"The video frame depicts a large, ominous dust cloud looming over a cityscape, which includes modern residential buildings and streets. The dust cloud is massive and appears to be moving towards the city, suggesting a severe dust storm. The text in the top left corner, which is in Korean, mentions the severity of the dust storm and includes the Chinese flag, indicating that the location is in China. The text translates to \"Dust storm in China, what are the regulations?\" There are parked cars and some greenery along the streets, but the dust cloud dominates the scene, creating a sense of impending danger or disruption.",
"The current frame shows a continuation of the previous scene, with the dust cloud still prominently visible, suggesting the ongoing impact of the dust storm. The text at the bottom, which is in Korean, mentions the large dust storm in China, highlighting the severity of the situation. The presence of the Chinese flag reaffirms the location. The atmosphere remains dominated by the dust cloud, indicating ongoing disruption or danger.",
"The current frame continues to depict the ongoing dust storm, as evidenced by the persistent dense dust cloud. The text at the bottom, still in Korean, now focuses on the impacts of the dust storm in China, emphasizing the scale of the event. The Chinese flag remains visible, reinforcing the geographical context. The scene maintains its depiction of the dust storm's disruptive presence.",
"The video features a news anchor in a studio with a large screen behind her displaying an image of a severe dust storm. The screen shows a dense, orange-hued dust cloud, indicating the ongoing nature of the dust storm. The text at the bottom of the screen is in Korean and discusses the impacts of the dust storm, particularly focusing on the environmental and possibly health-related consequences. The Chinese flag is visible, which suggests that the dust storm's origins or significant impact areas include China. The scene emphasizes the widespread and severe nature of the dust storm."
] |
Yddr-IKR_KA
|
A severe dust storm is affecting a city in China, causing disruption and danger. The storm is depicted as a massive, ominous dust cloud looming over the cityscape, with text in Korean discussing its severity and impacts. The scene transitions to a news studio where an anchor is reporting on the storm, displaying a large screen showing the dust storm's dense cloud and discussing its environmental and health consequences. The video focuses on the storm's scale and effects, with the Chinese flag indicating the geographical context. Overall, the video portrays the dust storm as a significant and potentially hazardous event affecting China.
|
The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.
|
[
"Strong winds blowing sand and dust across the Eastern Gobi desert steppe during the sandstorm on March 14, 2021",
"Reduced visibility on roads and highways due to heavy sand and dust in the air during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"People wearing masks and protective gear to shield themselves from the massive sandstorm that hit the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and other parts of East Asia on 14 March 2021",
"Sand and dust accumulating on buildings, streets, and vehicles during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"Flights and daily activities being disrupted due to a sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, and Korean Peninsula on March 14, 2021"
] |
[
"People wearing masks walking in low visibility conditions during a sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on 14 March 2021",
"Cars driving with headlights on during the day due to reduced visibility caused by the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on March 14, 2021",
"Sandstorm reducing visibility in urban areas of the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"Flights grounded or delayed due to sandstorm in East Asia on March 14, 2021",
"Residents sealing windows and doors to prevent sand entry during the 14 March 2021 sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia"
] |
[
"Sandstorm reducing visibility on roads and highways in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"People wearing masks to protect themselves from the sandstorm in East Asia on March 14, 2021",
"Trees and buildings covered in dust and sand during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on 14 March 2021",
"Cars and buses driving slowly due to low visibility during a sandstorm in the Eastern Gobi desert or East Asia on March 14 2021",
"Residents cleaning up sand and dust from streets and homes after the 14 March 2021 sandstorm in East Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%99%A9%EC%82%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects and progression of the sandstorm, such as reduced visibility, dust in the air, and people protecting themselves from the storm. These events are likely to be visualized in a video associated with the query, given the geographical scope and impact of the sandstorm.\n\nEVENTS:\n1. Dust and sand reducing visibility on roads and highways\n2. People wearing masks to protect themselves from the sandstorm\n3. Trees and buildings covered in dust and sand\n4. Cars and buses driving slowly due to low visibility\n5. Residents cleaning up sand and dust from streets and homes",
"en_article_excerpt": "The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.",
"en_article_url": "https://en.wikipedia.org/wiki/2021_East_Asia_sandstorm",
"event_category": "disasters",
"event_name": "2021_sandstorm",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the 2021 East Asia sandstorm, it makes sense to consider the typical visual consequences of a large-scale sandstorm, such as reduced visibility, strong winds, and massive amounts of dust and sand particles in the air. Events that could be visualized in a video would likely involve the storm's impact on the environment and daily life, rather than abstract or behind-the-scenes occurrences.\n\nEVENTS:\n1. Strong winds blowing sand and dust across the Eastern Gobi desert steppe\n2. Reduced visibility on roads and highways due to heavy sand and dust in the air\n3. People wearing masks and protective gear to shield themselves from the storm\n4. Sand and dust accumulating on buildings, streets, and vehicles\n5. Flights and daily activities being disrupted due to poor weather conditions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2021 East Asia sandstorm, it makes sense to extract events that are directly related to the effects and progression of the sandstorm, such as reduced visibility, transportation disruptions, and daily life impacts. These events should be concrete and visual, allowing them to be reasonably captioned in a video frame associated with the query.\n\nEVENTS:\n1. People wearing masks walking in low visibility conditions\n2. Cars driving with headlights on during the day due to reduced visibility\n3. Sandstorm reducing visibility in urban areas\n4. Flights being grounded or delayed at airports\n5. Residents sealing windows and doors to prevent sand entry",
"video_URL": "http://youtube.com/watch?v=3BbP2OHdePY",
"video_description": "뿌연 하루였습니다. 몽골에서 시작된 모래폭풍이 중국을 거쳐 한반도에 상륙한 건데, 내일은 전국에서 황사가루가 섞인 '흙비'까지 내릴 전망입니다. [Ch.19] 사실을 보고 진실을 말합니다. 👍🏻 공식 홈페이지 http://news.tvchosun.com/ 👍🏻 공식 페이스북 https://www.facebook.com/tvchosunnews/ 👍🏻 공식 트위터 https://twitter.com/TVChosunNews * 뉴스제보 : 이메일(tvchosun@chosun.com), 카카오톡(tv조선제보), 전화(1661-0190)",
"video_id": "3BbP2OHdePY"
}
| 4
|
[
"The image shows an urban street scene with a significant amount of dust or haze in the air, partially obscuring the view of tall buildings on either side. The text \"내일 화비 내린다\" is visible, indicating that a yellow dust storm will arrive the next day. The time displayed on the screen is 21:27, and there is additional text mentioning an event related to \"문화재청, 이건희 회장 소장 문화재 6월부터 국내외 전시,\" suggesting an upcoming exhibition of cultural assets from Lee Kun-hee's collection starting in June.",
"The current frame shows a view of a cityscape with a significant amount of dust or haze in the air, reducing visibility. The text \"내일 화비 내린다\" remains visible, continuing the message about the impending yellow dust storm. The time displayed is now 21:28. There is new text at the bottom of the screen indicating a wildfire situation, stating \"중국발 화재, 백령도 상륙한 뒤 서쪽 지역으로 확대,\" which translates to \"Wildfire from China, after landing on Baengnyeongdo, is spreading to western regions.\" This suggests an ongoing and expanding environmental crisis affecting the area.",
"The current frame captures a rainy urban street scene with several vehicles navigating the wet road. The headlights of cars and the rear lights of motorcycles are visible, indicating active traffic despite the weather. The text \"내일 화비 내린다\" is still displayed, suggesting the video is continuing to report on weather conditions. A lower text reads \"북쪽으로부터 남하한 한 공기가 인해 대기가 매우 불안정해지며 전국 곳곳에 비가 내리겠다고,\" which translates to \"The air coming from the north is causing unstable conditions, and rain is expected in various places across the country,\" further detailing the weather situation. The time stamp at the bottom indicates it is 21:28, maintaining consistency with the previous frame.",
"The current frame shows a farm field being affected by strong winds and possibly hail or rain, as evidenced by the bent and damaged plants. The text \"자료화면\" indicates this might be archival footage, suggesting the video is providing a historical perspective on the impact of harsh weather conditions on agriculture. The ongoing text \"북쪽으로부터 남하한 한 공기가 인해 대기가 매우 불안정해지며 전국 곳곳에 비가 내리겠다고,\" which translates to \"The air coming from the north is causing unstable conditions, and rain is expected in various places across the country,\" continues to emphasize the adverse weather effects. The timestamp remains at 21:28, keeping the frame consistent with the previous description in terms of time."
] |
3BbP2OHdePY
|
The video depicts an urban and environmental crisis scenario, set against the backdrop of severe weather conditions. It begins with a hazy cityscape, warning of an impending yellow dust storm and mentioning an unrelated cultural event. The scene progresses to show a wildfire spreading from China to western regions, indicating a significant environmental threat. The video then shifts to a rainy urban street with ongoing traffic, highlighting the challenges posed by unstable weather conditions affecting the entire country. Finally, it shows the impact of harsh weather on agriculture, with strong winds and possibly hail damaging farm fields. Throughout, the video maintains a consistent time frame of 21:27 to 21:28, suggesting a real-time reporting style, and emphasizes the severity of the weather situation and its widespread effects.
|
The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.
|
[
"Strong winds blowing sand and dust across the Eastern Gobi desert steppe during the sandstorm on March 14, 2021",
"Reduced visibility on roads and highways due to heavy sand and dust in the air during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"People wearing masks and protective gear to shield themselves from the massive sandstorm that hit the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and other parts of East Asia on 14 March 2021",
"Sand and dust accumulating on buildings, streets, and vehicles during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"Flights and daily activities being disrupted due to a sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, and Korean Peninsula on March 14, 2021"
] |
[
"People wearing masks walking in low visibility conditions during a sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on 14 March 2021",
"Cars driving with headlights on during the day due to reduced visibility caused by the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on March 14, 2021",
"Sandstorm reducing visibility in urban areas of the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"Flights grounded or delayed due to sandstorm in East Asia on March 14, 2021",
"Residents sealing windows and doors to prevent sand entry during the 14 March 2021 sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia"
] |
[
"Sandstorm reducing visibility on roads and highways in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"People wearing masks to protect themselves from the sandstorm in East Asia on March 14, 2021",
"Trees and buildings covered in dust and sand during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on 14 March 2021",
"Cars and buses driving slowly due to low visibility during a sandstorm in the Eastern Gobi desert or East Asia on March 14 2021",
"Residents cleaning up sand and dust from streets and homes after the 14 March 2021 sandstorm in East Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%99%A9%EC%82%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects and progression of the sandstorm, such as reduced visibility, dust in the air, and people protecting themselves from the storm. These events are likely to be visualized in a video associated with the query, given the geographical scope and impact of the sandstorm.\n\nEVENTS:\n1. Dust and sand reducing visibility on roads and highways\n2. People wearing masks to protect themselves from the sandstorm\n3. Trees and buildings covered in dust and sand\n4. Cars and buses driving slowly due to low visibility\n5. Residents cleaning up sand and dust from streets and homes",
"en_article_excerpt": "The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.",
"en_article_url": "https://en.wikipedia.org/wiki/2021_East_Asia_sandstorm",
"event_category": "disasters",
"event_name": "2021_sandstorm",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the 2021 East Asia sandstorm, it makes sense to consider the typical visual consequences of a large-scale sandstorm, such as reduced visibility, strong winds, and massive amounts of dust and sand particles in the air. Events that could be visualized in a video would likely involve the storm's impact on the environment and daily life, rather than abstract or behind-the-scenes occurrences.\n\nEVENTS:\n1. Strong winds blowing sand and dust across the Eastern Gobi desert steppe\n2. Reduced visibility on roads and highways due to heavy sand and dust in the air\n3. People wearing masks and protective gear to shield themselves from the storm\n4. Sand and dust accumulating on buildings, streets, and vehicles\n5. Flights and daily activities being disrupted due to poor weather conditions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2021 East Asia sandstorm, it makes sense to extract events that are directly related to the effects and progression of the sandstorm, such as reduced visibility, transportation disruptions, and daily life impacts. These events should be concrete and visual, allowing them to be reasonably captioned in a video frame associated with the query.\n\nEVENTS:\n1. People wearing masks walking in low visibility conditions\n2. Cars driving with headlights on during the day due to reduced visibility\n3. Sandstorm reducing visibility in urban areas\n4. Flights being grounded or delayed at airports\n5. Residents sealing windows and doors to prevent sand entry",
"video_URL": "http://youtube.com/watch?v=6sLn3wHQvCQ",
"video_description": "초대형 모래폭풍에 산사태까지…재난영화 방불케한 순간 포착 (서울=연합뉴스) 중국 북서부 간쑤성에서 초대형 모래폭풍이 도시를 집어삼키고 산사태로 엄청난 흙더미가 쓰나미처럼 쏟아져 내리는 등 흡사 재난 영화의 한 장면 같은 순간들이 카메라에 포착됐습니다. 지난 25일 간쑤성 과저우현에서는 높이 수십 미터에 달하는 거대한 모래폭풍이 도시를 덮쳤습니다. 모래폭풍은 마치 거대한 모래 장벽 같은 모습이었습니다. 모래폭풍이 시야를 가리고 해를 가리면서 일대는 오렌지색으로 물들었습니다. 가시거리가 50m 미만, 공기 질이 최악 수준으로 내려앉으면서 대기오염 최고단계인 황색경보가 발령됐습니다. 한편 같은 날 간쑤성 룽난시에서는 산사태가 발생했습니다. 산 중턱의 한 부분에서부터 흙더미와 나무가 쓰나미처럼 산 아래로 쏟아져 내리면서 도로에 있던 관광객 수십 명이 비명을 지르며 도망쳤습니다. 이날 산사태로 차량 한 대가 파손됐지만 인명 피해는 없었다고 합니다. 영상으로 보시죠. 제작 : 김건태·안창주 영상 : 로이터 #연합뉴스 #중국 #황사 ◆ 연합뉴스 유튜브 : https://www.youtube.com/yonhap ◆ 연합뉴스 홈페이지→ http://www.yna.co.kr/ ◆ 연합뉴스 페이스북→ https://www.facebook.com/yonhap/ ◆ 연합뉴스 인스타 : https://goo.gl/UbqiQb ◆ 연합뉴스 비디오메타 채널https://www.youtube.com/channel/UCTMCrbtHU0M0SR6TuBrL4Pw",
"video_id": "6sLn3wHQvCQ"
}
| 4
|
[
"A large dust storm is approaching a city, with buildings visible at the bottom of the image. The text \"연합뉴스\" indicates a news source, and the phrase \"시커멓고 거대한\" translates to \"dark and gigantic,\" describing the storm.",
"The dust storm now covers the city, significantly reducing visibility. The text \"공기질은 최악\" translates to \"air quality is the worst,\" indicating the severe impact on air quality. The storm continues to engulf the buildings and streets, emphasizing the extent of the situation.",
"A tree is engulfed in flames, with smoke billowing out. The text \"연합뉴스 (도망가 뛰리)\" translates to \"Yonhap News (Running away).\" The fire is spreading among the trees, with people visible in the foreground, likely reacting to the fire and smoke.",
"In the current frame, a person is depicted pointing upwards with a determined expression. The text \"구독하기\" translates to \"Subscribe,\" suggesting a call to action for viewers to subscribe to the content. The image features a cartoon character pointing towards a circular emblem with the text \"연합뉴스\" (Yonhap News)."
] |
6sLn3wHQvCQ
|
A severe dust storm hits a city, causing extremely poor air quality and significantly reducing visibility as it engulfs buildings and streets. The situation escalates with a fire breaking out, spreading among trees and prompting people to react. The video then shifts to a promotional image featuring a cartoon character pointing to the Yonhap News logo, suggesting the footage was part of a news report from Yonhap News.
|
The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.
|
[
"Strong winds blowing sand and dust across the Eastern Gobi desert steppe during the sandstorm on March 14, 2021",
"Reduced visibility on roads and highways due to heavy sand and dust in the air during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"People wearing masks and protective gear to shield themselves from the massive sandstorm that hit the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and other parts of East Asia on 14 March 2021",
"Sand and dust accumulating on buildings, streets, and vehicles during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"Flights and daily activities being disrupted due to a sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, and Korean Peninsula on March 14, 2021"
] |
[
"People wearing masks walking in low visibility conditions during a sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on 14 March 2021",
"Cars driving with headlights on during the day due to reduced visibility caused by the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on March 14, 2021",
"Sandstorm reducing visibility in urban areas of the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"Flights grounded or delayed due to sandstorm in East Asia on March 14, 2021",
"Residents sealing windows and doors to prevent sand entry during the 14 March 2021 sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia"
] |
[
"Sandstorm reducing visibility on roads and highways in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, and East Asia on 14 March 2021",
"People wearing masks to protect themselves from the sandstorm in East Asia on March 14, 2021",
"Trees and buildings covered in dust and sand during the sandstorm in the Eastern Gobi desert, Mongolian Plateau, Loess Plateau, North China Plain, Korean Peninsula, or East Asia on 14 March 2021",
"Cars and buses driving slowly due to low visibility during a sandstorm in the Eastern Gobi desert or East Asia on March 14 2021",
"Residents cleaning up sand and dust from streets and homes after the 14 March 2021 sandstorm in East Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/%ED%99%A9%EC%82%AC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects and progression of the sandstorm, such as reduced visibility, dust in the air, and people protecting themselves from the storm. These events are likely to be visualized in a video associated with the query, given the geographical scope and impact of the sandstorm.\n\nEVENTS:\n1. Dust and sand reducing visibility on roads and highways\n2. People wearing masks to protect themselves from the sandstorm\n3. Trees and buildings covered in dust and sand\n4. Cars and buses driving slowly due to low visibility\n5. Residents cleaning up sand and dust from streets and homes",
"en_article_excerpt": "The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.",
"en_article_url": "https://en.wikipedia.org/wiki/2021_East_Asia_sandstorm",
"event_category": "disasters",
"event_name": "2021_sandstorm",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: To extract events related to the 2021 East Asia sandstorm, it makes sense to consider the typical visual consequences of a large-scale sandstorm, such as reduced visibility, strong winds, and massive amounts of dust and sand particles in the air. Events that could be visualized in a video would likely involve the storm's impact on the environment and daily life, rather than abstract or behind-the-scenes occurrences.\n\nEVENTS:\n1. Strong winds blowing sand and dust across the Eastern Gobi desert steppe\n2. Reduced visibility on roads and highways due to heavy sand and dust in the air\n3. People wearing masks and protective gear to shield themselves from the storm\n4. Sand and dust accumulating on buildings, streets, and vehicles\n5. Flights and daily activities being disrupted due to poor weather conditions",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2021 East Asia sandstorm was a meteorological phenomenon that began in the Eastern Gobi desert steppe on March 14, and subsequently spread to the entire Mongolian Plateau South, the Loess Plateau, the North China Plain and the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about the 2021 East Asia sandstorm, it makes sense to extract events that are directly related to the effects and progression of the sandstorm, such as reduced visibility, transportation disruptions, and daily life impacts. These events should be concrete and visual, allowing them to be reasonably captioned in a video frame associated with the query.\n\nEVENTS:\n1. People wearing masks walking in low visibility conditions\n2. Cars driving with headlights on during the day due to reduced visibility\n3. Sandstorm reducing visibility in urban areas\n4. Flights being grounded or delayed at airports\n5. Residents sealing windows and doors to prevent sand entry",
"video_URL": "http://youtube.com/watch?v=qaan1wMGnN0",
"video_description": "거대한 '모래 폭풍'이 하늘을 뒤덮었습니다. 지난 주말 몽골의 건조한 초원에서 발원한 '황사 바람'입니다. 태풍급 강풍을 동반한 모래 먼지가 한적한 작은 마을을 집어삼킬 듯한 기세입니다. 초원은 물론이고 주변 도시들도 앞을 분간하기 어려운 먼지 속에 갇혔습니다. '황색 폭풍'의 습격으로 몽골 초원지대 유목민 약 600명이 한때 실종됐습니다. 다행히 대부분 살아서 돌아왔지만 5살 어린 아이 를 포함해 10여 명은 결국 숨졌습니다. 최고 풍속 초속 30m가 넘는 모래 폭풍에 10층 높이의 철탑도 맥없이 부러졌습니다. 인근 지역 주민들은 모래 바람의 공포 속에, 정전 까지 돼 암흑과 추위에 떨었습니다. 몽골 초원과 가까운 중국 북쪽 네이멍구에서는 모래가 섞인 눈보라가 몰아쳤습니다. 강풍에 밀려온 모래와 눈이 민가를 뒤덮으면서 집안에 갇히는 주민들도 속출했습니다. 경찰 대여섯 명이 삽과 수레로 눈 모래를 걷어낸 지 2시간 만에 71살 할머니가 구조되기도 했습니다. 이번 황사는 강한 바람이 모래 먼지를 멀리까지 실어나르면서 '10년 만의 황사'로 강해졌습니다. 특히 몽골 초원의 봄풀이 아직 자라기 전이어서 흙먼지가 많았다고 중국 기상 당국은 분석했습니다. 황사로 뒤덮였던 베이징에는 다시 맑은 날씨가 찾아왔습니다. 강한 바람 때문에 모래 먼지는 하루 만에 사라졌 는데, 이번 봄에 황사가 몇 차례 더 예보가 돼있는 상황입니다. 취재기자 : 강성웅 촬영편집 : 고광 자막뉴스 : 윤현경 ▶ 기사 원문 : https://www.ytn.co.kr/_ln/0134_202103170936125411 ▶ 제보 하기 : https://mj.ytn.co.kr/mj/mj_write.php ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN & YTN plus 무단 전재 및 재배포금지",
"video_id": "qaan1wMGnN0"
}
| 4
|
[
"The frame shows barbed wire against an orange background, likely indicating a barrier or boundary. Korean text at the bottom provides context, mentioning that about 600 Ukrainian civilians were killed when a Russian missile struck a shopping mall during a \"powerful explosion.\" The source \"YTN\" is visible in the top right corner.",
"The frame depicts a heavily foggy street scene, with visibility significantly reduced. The traffic light is visible but its signals are difficult to discern due to the fog. Korean text at the bottom indicates that dust and smog have descended on the northern part of China, particularly in Beijing. The source \"YTN\" is again visible in the top right corner, suggesting a continuation of the news coverage from the previous frame.",
"The frame depicts a heavily dusty street scene with a significant decrease in visibility. Vehicles, including a bus and a car, are seen navigating through the dense dust, emphasizing the challenging conditions. Korean text at the bottom translates to a description of the event, mentioning that the dust has turned the sky yellow and lasted for about 10 minutes. The source \"YTN\" is visible in the top right corner, indicating ongoing news coverage.",
"The current frame continues to depict the heavily dusty street scene with ongoing challenging visibility. The vehicles, including the bus and car from the previous frame, are still navigating through the dense dust. The Korean text at the bottom is no longer visible, but the source \"YTN\" remains in the top right corner, indicating the news coverage is still ongoing. The dust appears to have slightly cleared compared to the previous frame, suggesting a gradual improvement in the conditions."
] |
qaan1wMGnN0
|
A news video covers two separate events. The first event reports on a Russian missile strike that killed approximately 600 Ukrainian civilians at a shopping mall. The second event shows the effects of a severe dust and smog phenomenon in northern China, particularly in Beijing, which reduces visibility and causes hazardous conditions for vehicles navigating the streets. The dust storm lasts for about 10 minutes, turning the sky yellow, before gradually clearing. The video is sourced from "YTN" and presents these events in a serious news context.
|
Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.
|
[
"People sweating and fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Temperature readings on thermometers exceeding record highs during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens gathering at public water parks and pools to cool off during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Ambulances rushing to treat heat stroke patients during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Farmers inspecting withered and dried-out crops due to the heat wave in Northeast Asia during July 2018"
] |
[
"People sweating and fanning themselves in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Temperature readings displayed on outdoor thermometers showing extremely high numbers during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queueing for water or other cooling supplies during the July 2018 heat wave in Japan, North Korea, South Korea, and China",
"Heat-related illnesses being treated at hospitals or medical tents during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Dried-up or wilted plants and crops due to the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia"
] |
[
"People fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queuing for water in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"thermometer screens displaying extremely high temperatures during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Trees and plants wilting due to the heat wave in Japan, North Korea, South Korea, China, and Northeast Asia in July 2018",
"People wearing masks and hats to protect themselves from the sun during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2018%EB%85%84_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%8F%AD%EC%97%BC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are commonly associated with extreme heat and its effects on people and the environment. These events should be visual and tangible, allowing them to be captured in a video, such as people trying to cool themselves, impacts on daily life, or consequences of the heat wave.\n\nEVENTS:\n1. People fanning themselves on the streets\n2. Citizens queuing for water in public spaces\n3. thermometer screens displaying extremely high temperatures\n4. Trees and plants wilting due to the heat\n5. People wearing masks and hats to protect themselves from the sun",
"en_article_excerpt": "Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.",
"en_article_url": "https://en.wikipedia.org/wiki/2018_Northeast_Asia_heat_wave",
"event_category": "disasters",
"event_name": "2018_northeast_asia_heat_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a heat wave, such as extreme weather conditions, people trying to cool themselves, and potential consequences of the heat wave. These events should be concrete and visual, allowing them to be easily depicted in a video.\n\nEVENTS:\n1. People sweating and fanning themselves on the streets\n2. Temperature readings on thermometers exceeding record highs\n3. Citizens gathering at public water parks and pools to cool off\n4. Ambulances rushing to treat heat stroke patients\n5. Farmers inspecting withered and dried-out crops due to the heat wave",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are visually representative of extreme heat and its effects on daily life and the environment. These events should be common consequences of a heat wave and plausible to be captured in a video during that time in the mentioned regions.\n\nEVENTS:\n1. People sweating and fanning themselves in public spaces\n2. Temperature readings displayed on outdoor thermometers showing extremely high numbers\n3. Citizens queueing for water or other cooling supplies\n4. Heat-related illnesses being treated at hospitals or medical tents\n5. Dried-up or wilted plants and crops due to the heat wave",
"video_URL": "http://youtube.com/watch?v=0ZROpcl5nUE",
"video_description": "열기가 식을 새도 없이 계속 쌓여만 가고 있습니다. 폭염 특보도 확대, 강화되고 있는데요. 오늘 오전 11시를 기해 경기 남부와 강원 영서 남부에도 폭염경보가 내려졌고, 전국 대부분 지역으로 폭염주의보가 확대됐습니다. 현재 경북 경주 기온이 36.6도로 전국에서 가장 뜨겁고요. 대구 36.4도, 서울도 32.9도까지 올라 올해 최고 기온을 기록할 것으로 보입니다. 밤에도 열기는 쉽게 식지 못하고 열대야로 이어지는 곳이 많겠는데요. 내일 서울의 아침 기온 24도, 대구 25도로 오늘 밤사이에도 더위로 잠 못 이루는 곳이 많겠고요. 낮에는 다시 서울 기온이 33도, 대구 37도까지 오르는 등 오늘과 비슷하겠습니다. 현재 전국에 맑은 하늘이 드러나 있는데요. 내일도 종일 쾌청한 하늘이 함께 하겠지만, 강한 자외선에는 주의하셔야겠습니다. 당분간 뚜렷한 비 소식 없이 낮에는 폭염이, 밤에는 열대야가 기승을 부리겠습니다. 지금까지 날씨 정보였습니다. ▶ 기사 원문 : http://www.ytn.co.kr/_ln/0108_201807151449122225 ▶ 제보 안내 : http://goo.gl/gEvsAL, 모바일앱, 8585@ytn.co.kr, #2424 ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ [ 한국 뉴스 채널 와이티엔 / Korea News Channel YTN ]",
"video_id": "0ZROpcl5nUE"
}
| 4
|
[
"In the video, a woman is presenting weather information, with a map highlighting different regions in South Korea. The map shows areas marked in orange and red, indicating varying weather conditions, possibly temperatures or air quality levels. The background depicts a sunny beach scene with blue skies. The text on the screen includes \"폭염특보\" (heatwave warning) and \"폭염주의보\" (heatwave advisory), suggesting that the weather forecast is focused on the hot conditions across the region.",
"The video continues to feature the weather presenter, now with a chart displaying temperature data. The chart shows that the nighttime temperature in Seoul is expected to be 24°C, while the daytime temperature will reach 33°C. In contrast, the nighttime temperature for the country is 25°C, and the daytime temperature is projected to be 37°C. The chart is accompanied by icons of the sun and a crescent moon, indicating the times for the temperatures. The background remains a sunny sky with some clouds, consistent with the theme of the weather segment. The text in the chart reads \"내일도 폭염 계속\" (Extreme heat continues tomorrow), reinforcing the forecast of continued high temperatures.",
"The weather presenter continues her segment, now standing in front of a map showing weather patterns over East Asia. She is wearing a pink dress and appears to be speaking, likely discussing the weather conditions depicted on the map. The map shows cloud cover and possibly weather fronts affecting the region, with a focus on Korea and surrounding areas. The presence of the map suggests she is providing detailed information on current and upcoming weather conditions, potentially addressing recent or forecasted weather events like typhoons or storms in the East Asian region. The background remains consistent with the previous frames, featuring a bright and clear sky, which aligns with the theme of the weather forecast.",
"The weather presenter, still in her pink dress, is now standing in front of a weather forecast display that provides a detailed week-by-week outlook for Seoul. The display shows sunny conditions with high temperatures ranging from 32 to 33 degrees Celsius and low temperatures from 24 to 25 degrees Celsius, indicating a consistent warm and sunny pattern throughout the week. The background remains the same, with a bright and clear sky, reinforcing the sunny weather forecast. This frame suggests the presenter is discussing upcoming weather conditions, providing viewers with information on temperatures and weather patterns for the week ahead."
] |
0ZROpcl5nUE
|
A woman presents a weather forecast for South Korea, highlighting a heatwave warning and advisory across the region. She displays a map showing varying weather conditions and a chart with temperature data, indicating high temperatures during the day and relatively warm nights, with Seoul's temperatures ranging from 24°C to 33°C. The presenter then discusses weather patterns over East Asia, potentially addressing storms or typhoons, before providing a week-by-week outlook for Seoul, forecasting consistent warm and sunny conditions with temperatures between 24°C and 33°C. The overall context is a weather segment focused on the hot conditions and upcoming weather events in South Korea and the East Asian region.
|
Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.
|
[
"People sweating and fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Temperature readings on thermometers exceeding record highs during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens gathering at public water parks and pools to cool off during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Ambulances rushing to treat heat stroke patients during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Farmers inspecting withered and dried-out crops due to the heat wave in Northeast Asia during July 2018"
] |
[
"People sweating and fanning themselves in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Temperature readings displayed on outdoor thermometers showing extremely high numbers during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queueing for water or other cooling supplies during the July 2018 heat wave in Japan, North Korea, South Korea, and China",
"Heat-related illnesses being treated at hospitals or medical tents during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Dried-up or wilted plants and crops due to the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia"
] |
[
"People fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queuing for water in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"thermometer screens displaying extremely high temperatures during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Trees and plants wilting due to the heat wave in Japan, North Korea, South Korea, China, and Northeast Asia in July 2018",
"People wearing masks and hats to protect themselves from the sun during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2018%EB%85%84_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%8F%AD%EC%97%BC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are commonly associated with extreme heat and its effects on people and the environment. These events should be visual and tangible, allowing them to be captured in a video, such as people trying to cool themselves, impacts on daily life, or consequences of the heat wave.\n\nEVENTS:\n1. People fanning themselves on the streets\n2. Citizens queuing for water in public spaces\n3. thermometer screens displaying extremely high temperatures\n4. Trees and plants wilting due to the heat\n5. People wearing masks and hats to protect themselves from the sun",
"en_article_excerpt": "Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.",
"en_article_url": "https://en.wikipedia.org/wiki/2018_Northeast_Asia_heat_wave",
"event_category": "disasters",
"event_name": "2018_northeast_asia_heat_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a heat wave, such as extreme weather conditions, people trying to cool themselves, and potential consequences of the heat wave. These events should be concrete and visual, allowing them to be easily depicted in a video.\n\nEVENTS:\n1. People sweating and fanning themselves on the streets\n2. Temperature readings on thermometers exceeding record highs\n3. Citizens gathering at public water parks and pools to cool off\n4. Ambulances rushing to treat heat stroke patients\n5. Farmers inspecting withered and dried-out crops due to the heat wave",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are visually representative of extreme heat and its effects on daily life and the environment. These events should be common consequences of a heat wave and plausible to be captured in a video during that time in the mentioned regions.\n\nEVENTS:\n1. People sweating and fanning themselves in public spaces\n2. Temperature readings displayed on outdoor thermometers showing extremely high numbers\n3. Citizens queueing for water or other cooling supplies\n4. Heat-related illnesses being treated at hospitals or medical tents\n5. Dried-up or wilted plants and crops due to the heat wave",
"video_URL": "http://youtube.com/watch?v=-HJV2kcDUAA",
"video_description": "서울 기온이 39도를 넘어 관측 사상 최고 기록을 갈아치운 1일 오후 서울 지하철역에서는 전력에 과부하가 걸려 스크린도어에 이상이 생겼다. 서울교통공사에 따르면 이날 오후 2시 39분께 지하철 4호선 동작역에서 당고개 방향 승강장 안전문(스크린도어)이 작동하지 않는 현상이 발생했다. 공사 측은 \"가정에서 전력에 이상이 생기면 화재 등 사고를 예방하기 위해 두꺼비집(누전차단기)이 내려가듯이, 승강장 쪽 전력에서 같은 작용이 일어난 것\"이라고 설명했다. 동작역 스크린도어는 오후 5시 30분 현재도 열려있는 상태로, 승강장에는 안전요원이 배치돼 만일의 사고에 대비하고 있다. 스크린도어가 개방된 상태일 뿐, 열차 운행에는 지장이 없는 상황이다. 공사 관계자는 \"밤에 운행을 마친 뒤에 정확한 원인을 조사할 방침\"이라면서 \"고온으로 인해 일부 부품에 문제가 생겼을 가능성도 염두에 두고 있다\"고 말했다. 영상취재 - [연합뉴스TV 차영우] ◆ 연합뉴스 홈페이지→ http://www.yna.co.kr/ ◆ 이 시각 많이 본 기사 → https://goo.gl/VQTsSZ ◆ 오늘의 핫뉴스 → https://goo.gl/WyGXpG ◆ 현장영상 → https://goo.gl/5aZcx8 ◆ 카드뉴스 →https://goo.gl/QKfDTH ◆연합뉴스 공식 SNS◆ ◇페이스북→ https://www.facebook.com/yonhap/ ▣ 연합뉴스 유튜브 채널 구독 : https://goo.gl/pL7TmT",
"video_id": "-HJV2kcDUAA"
}
| 4
|
[
"The current frame shows a subway train at a station platform. Passengers are visible inside the train through the windows, and one person appears to be using a mobile device. A sign above the door indicates the next stop as \"Transfer\" with an arrow pointing left. There is also a warning sign on the platform advising caution when boarding. A map on the right shows the subway routes and stations. The environment suggests a typical urban transit setting.",
"The current frame shows the interior of a subway train with passengers seated and standing inside. The lighting is bright, indicating the train is well-lit. The passengers appear to be engaged in various activities, some looking out the windows, and others focusing on their phones or personal belongings. The environment suggests a typical urban transit setting, with the train stationary, likely at a station. The focus is on the passengers and their interactions within the train.",
"The video continues with the focus shifting to a close-up of a subway train's control panel, highlighting a red indicator light and a yellow indicator light, suggesting the train is undergoing some form of maintenance or inspection. The control panel displays the text \"장애\" and \"체열림\" (indicating an alert or malfunction in the system), indicating potential issues with the train. The environment appears to be a maintenance or inspection area, with a focus on the technical details of the train. The lighting remains bright, maintaining the clarity of the scene. The scene suggests that the train is stationary, likely at a station, but with a focus on the technical aspects of the train's operation.",
"The scene transitions to a close-up of a subway train’s doors, with a focus on the transparent sections and the structural elements of the door mechanism. The lighting is bright, enhancing the visibility of the door components. The environment continues to suggest a maintenance or inspection setting, with the train still stationary. The red indicator lights visible in the frame suggest the train is undergoing some form of maintenance or inspection. The text \"연합뉴스\" (Yonhap News Agency) at the top right indicates that this is a news report."
] |
-HJV2kcDUAA
|
A subway train is depicted in various settings, initially at a station platform with passengers on board, then inside the train with passengers engaged in different activities. The scene shifts to a maintenance or inspection area, focusing on the train's control panel, which indicates potential issues with the system, and then to a close-up of the train's doors, highlighting its mechanism. The video suggests a news report, possibly covering a technical issue or maintenance operation of the subway train, capturing both the passenger experience and the technical aspects of the train's operation in an urban transit setting.
|
Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.
|
[
"People sweating and fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Temperature readings on thermometers exceeding record highs during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens gathering at public water parks and pools to cool off during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Ambulances rushing to treat heat stroke patients during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Farmers inspecting withered and dried-out crops due to the heat wave in Northeast Asia during July 2018"
] |
[
"People sweating and fanning themselves in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Temperature readings displayed on outdoor thermometers showing extremely high numbers during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queueing for water or other cooling supplies during the July 2018 heat wave in Japan, North Korea, South Korea, and China",
"Heat-related illnesses being treated at hospitals or medical tents during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Dried-up or wilted plants and crops due to the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia"
] |
[
"People fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queuing for water in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"thermometer screens displaying extremely high temperatures during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Trees and plants wilting due to the heat wave in Japan, North Korea, South Korea, China, and Northeast Asia in July 2018",
"People wearing masks and hats to protect themselves from the sun during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2018%EB%85%84_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%8F%AD%EC%97%BC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are commonly associated with extreme heat and its effects on people and the environment. These events should be visual and tangible, allowing them to be captured in a video, such as people trying to cool themselves, impacts on daily life, or consequences of the heat wave.\n\nEVENTS:\n1. People fanning themselves on the streets\n2. Citizens queuing for water in public spaces\n3. thermometer screens displaying extremely high temperatures\n4. Trees and plants wilting due to the heat\n5. People wearing masks and hats to protect themselves from the sun",
"en_article_excerpt": "Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.",
"en_article_url": "https://en.wikipedia.org/wiki/2018_Northeast_Asia_heat_wave",
"event_category": "disasters",
"event_name": "2018_northeast_asia_heat_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a heat wave, such as extreme weather conditions, people trying to cool themselves, and potential consequences of the heat wave. These events should be concrete and visual, allowing them to be easily depicted in a video.\n\nEVENTS:\n1. People sweating and fanning themselves on the streets\n2. Temperature readings on thermometers exceeding record highs\n3. Citizens gathering at public water parks and pools to cool off\n4. Ambulances rushing to treat heat stroke patients\n5. Farmers inspecting withered and dried-out crops due to the heat wave",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are visually representative of extreme heat and its effects on daily life and the environment. These events should be common consequences of a heat wave and plausible to be captured in a video during that time in the mentioned regions.\n\nEVENTS:\n1. People sweating and fanning themselves in public spaces\n2. Temperature readings displayed on outdoor thermometers showing extremely high numbers\n3. Citizens queueing for water or other cooling supplies\n4. Heat-related illnesses being treated at hospitals or medical tents\n5. Dried-up or wilted plants and crops due to the heat wave",
"video_URL": "http://youtube.com/watch?v=MZR5vPcBcI4",
"video_description": "극한 폭염속에도 경남 밀양 얼음골에선 주먹만한 얼음 기둥이 관측됐습니다. 신기한 건 이례적인 더위 속에서도 이 얼음이 평년보다 더 오랫동안, 꽁꽁 얼어있다는건데요. 어떻게 된 일일까요. 최진석 기자가 소개합니다. [리포트] 경남 밀양 재약산 해발 600m. 불볕더위에 바깥 온도는 36도를 넘어섰지만 바위 사이로 시원한 바람이 나옵니다. 바위 안쪽 온도를 재보니 서늘한 가을 기온인 9.2도. 바위 틈엔 주먹만한 얼음 기둥까지 있습니다. 이 지역의 명소이자 천연기념물 224호인 얼음골입니다. 불볕더위에 지친 사람들은 시원한 바람이 신기하기만 합니다. [이미경/부산광역시 북구 : \"여기까지 올라오는데 참 많이 힘들었는데, 너무 선선한 바람이 부니까 너무 너무 좋아요.\"] 바위 틈에서는 에어컨을 틀어놓은 것처럼 차가운 자연 바람이 나오고 있습니다. 열화상카메라로 온도를 재보니 1도밖에 되지 않습니다. 이곳의 얼음은 보통 장마철인 6,7월이면 다 녹아 없어지지만 올해는 계속되는 폭염의 상황에도 더 꽁꽁 얼어붙은 채 남아있습니다. [변희룡/부경대 지구환경과학과 명예교수 : \"10여 년 관측했는데, 올해 얼음이 가장 오래 관측됐거든요. 장마철에 비가 많이 오면씻겨지는데 장마철에 비가 많이 안왔어요.\"] 기록적인 폭염 속에 더위를 식히려 지난달에만 2만 여명이 찾은 밀양 얼음골. 신비로운 비밀을 간직한 채 더위에 지친 이들의 색다른 피서지가 되고 있습니다. KBS 뉴스 최진석입니다.",
"video_id": "MZR5vPcBcI4"
}
| 4
|
[
"The image is a screen from a news broadcast on KBS1 featuring Korean text. The text at the top reads \"일음골, 폭염에 더 '꽁꽁'\" which translates to \"Ileumgol, even more 'frozen' due to heatwave.\" The text at the bottom says \"경남 밀양시 산내면,\" translating to \"Gyeongnam Milyang City, Sanneom.\" The background shows a bright, overexposed sky with sunlight.",
"The current frame shows a group of four people standing outdoors with a backdrop of rocky terrain and greenery. They are posing for the camera, appearing to be tourists or visitors. The text at the bottom of the frame reads \"이미경 부산광역시 북구 여기까지 올라오는 데 참 많이 힘들었는데, 너무 선선한 바람이 부니깐 너무너무 좋았어요.\" which translates to \"Lee Mi-kyung from Busan Metropolitan City, Buk-gu, said that it was really difficult to climb up to here, but the cool breeze was so refreshing.\" The scene is likely part of a news segment related to the heatwave and the cooler temperatures at higher elevations, as suggested by the previous frame's text.",
"The current frame shows a close-up of a large icicle hanging from a rocky surface in what appears to be a natural cave or cave-like setting. The icicle is prominently featured, with water dripping from its bottom. The surrounding rocks have moss and other natural elements, suggesting a damp and cool environment. The text at the top of the frame reads \"KBS1\" and \"일읍골, 폭염에 더 '꽁꽁'\" which translates to \"Yeongeup Valley, even more 'freezing' in the heatwave.\" The image likely continues the narrative of contrasting environments, highlighting the coolness of Yeongeup Valley in the context of a heatwave.",
"The current frame continues to show a close-up view of the large icicle in the cave-like setting. The icicle remains prominently featured with water dripping from its bottom, reinforcing the cool environment contrasted with the external heatwave. The text \"KBS1\" is visible at the top of the frame, maintaining the narrative focus on Yeongeup Valley and its \"freezing\" conditions during the heatwave."
] |
MZR5vPcBcI4
|
A news segment on KBS1 reports on the contrasting environments during a heatwave, focusing on Yeongeup Valley in Gyeongnam Milyang City. The video shows tourists, including Lee Mi-kyung from Busan, visiting the area and appreciating the cool breeze after a difficult climb. The segment highlights the valley's uniquely cool conditions, featuring close-up shots of a large icicle in a natural cave-like setting, with water dripping from it, emphasizing the refreshing environment amidst the heatwave.
|
Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.
|
[
"People sweating and fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Temperature readings on thermometers exceeding record highs during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens gathering at public water parks and pools to cool off during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Ambulances rushing to treat heat stroke patients during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Farmers inspecting withered and dried-out crops due to the heat wave in Northeast Asia during July 2018"
] |
[
"People sweating and fanning themselves in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Temperature readings displayed on outdoor thermometers showing extremely high numbers during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queueing for water or other cooling supplies during the July 2018 heat wave in Japan, North Korea, South Korea, and China",
"Heat-related illnesses being treated at hospitals or medical tents during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Dried-up or wilted plants and crops due to the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia"
] |
[
"People fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queuing for water in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"thermometer screens displaying extremely high temperatures during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Trees and plants wilting due to the heat wave in Japan, North Korea, South Korea, China, and Northeast Asia in July 2018",
"People wearing masks and hats to protect themselves from the sun during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2018%EB%85%84_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%8F%AD%EC%97%BC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are commonly associated with extreme heat and its effects on people and the environment. These events should be visual and tangible, allowing them to be captured in a video, such as people trying to cool themselves, impacts on daily life, or consequences of the heat wave.\n\nEVENTS:\n1. People fanning themselves on the streets\n2. Citizens queuing for water in public spaces\n3. thermometer screens displaying extremely high temperatures\n4. Trees and plants wilting due to the heat\n5. People wearing masks and hats to protect themselves from the sun",
"en_article_excerpt": "Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.",
"en_article_url": "https://en.wikipedia.org/wiki/2018_Northeast_Asia_heat_wave",
"event_category": "disasters",
"event_name": "2018_northeast_asia_heat_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a heat wave, such as extreme weather conditions, people trying to cool themselves, and potential consequences of the heat wave. These events should be concrete and visual, allowing them to be easily depicted in a video.\n\nEVENTS:\n1. People sweating and fanning themselves on the streets\n2. Temperature readings on thermometers exceeding record highs\n3. Citizens gathering at public water parks and pools to cool off\n4. Ambulances rushing to treat heat stroke patients\n5. Farmers inspecting withered and dried-out crops due to the heat wave",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are visually representative of extreme heat and its effects on daily life and the environment. These events should be common consequences of a heat wave and plausible to be captured in a video during that time in the mentioned regions.\n\nEVENTS:\n1. People sweating and fanning themselves in public spaces\n2. Temperature readings displayed on outdoor thermometers showing extremely high numbers\n3. Citizens queueing for water or other cooling supplies\n4. Heat-related illnesses being treated at hospitals or medical tents\n5. Dried-up or wilted plants and crops due to the heat wave",
"video_URL": "http://youtube.com/watch?v=HCV2Hx8B0-c",
"video_description": "8월의 첫날인 오늘 최악의 폭염이 이어졌습니다. 오늘 얼마나 더웠는지 기온만 봐도 알 수가 있는데요. 강원도 홍천은 낮 기온 41도, 서울 39.6도까지 오르면서 기상 관측 이래 가장 높은 기온을 기록했습니다. 내일도 극심한 폭염은 계속됩니다. 해안가를 제외하고는 전국이 낮 기온 35도를 넘어서면서 뜨겁겠습니다. 서울과 춘천 39도, 대전과 대구는 38도까지 오르면서 오늘만큼 덥겠습니다. 토요일부터는 기온이 조금 내려가겠지만 낮 기온 35도 안팎으로 찜통더위는 이어집니다. 절기상 입추인 다음 주 화요일에도 낮 기온 35도로 무덥겠습니다. 현재 경기와 충남, 전북에는 오존주의보가 발효 중입니다. 내일도 전국이 맑겠고 자외선이 강해서 오존 농도가 높아질 수 있겠는데요. 중부와 전북 지역에서는 오존을 주의하셔야겠습니다. 아침 기온은 서울 29도로 밤 사이에도 무덥겠습니다. 한낮에는 광주 37도, 부산 35도로 오늘만큼 덥겠습니다. 바다의 물결은 제주해상에서 최고 4미터까지 거세게 일겠습니다. 더위가 밤낮을 가리지 않는 만큼 온열 질환에 걸리지 않도록 건강 관리를 잘해주시기 바랍니다. 날씨정보 전해드렸습니다.",
"video_id": "HCV2Hx8B0-c"
}
| 4
|
[
"The image shows a weather forecast presentation featuring a woman standing next to a graph depicting temperature trends. The graph includes red and orange lines indicating temperature changes, with text in Korean, including \"당분간 더위는 이어져\" (It will continue to be hot for a while) and temperature readings in Celsius. The woman is dressed in a white dress and appears to be providing information about the weather.",
"The current frame shows a weather forecast presentation featuring a woman standing next to a graph depicting temperature trends. The graph shows a downward trend in both high and low temperatures, with the high temperatures starting at 39.6°C and gradually decreasing to 35°C over a week. The low temperatures start at 27.8°C and decrease to 25°C. The woman, dressed in a white dress, appears to be providing an update on the weather, indicating a cooling trend in the upcoming days.",
"The weather forecast presentation continues with the woman still standing next to the graph. The graph still shows the downward trend in both high and low temperatures. The high temperatures remain at 35°C, while the low temperatures are now at 25°C. The woman, dressed in a white dress, continues to provide an update on the weather, emphasizing the cooling trend for the upcoming days.",
"The weather forecast presentation continues with the woman standing next to the graph. The graph still shows the downward trend in both high and low temperatures. The high temperatures remain at 35°C, while the low temperatures are now at 25°C. The woman, dressed in a white dress, continues to provide an update on the weather, emphasizing the cooling trend for the upcoming days."
] |
HCV2Hx8B0-c
|
A woman, dressed in a white dress, presents a weather forecast, standing next to a graph that displays temperature trends. The graph initially shows high temperatures of 39.6°C and low temperatures of 27.8°C, with a downward trend indicating a cooling pattern over the course of a week, eventually reaching highs of 35°C and lows of 25°C. Throughout the presentation, the woman provides updates on the weather, emphasizing that the hot weather will continue for a while but will eventually cool down in the upcoming days.
|
Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.
|
[
"People sweating and fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Temperature readings on thermometers exceeding record highs during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens gathering at public water parks and pools to cool off during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Ambulances rushing to treat heat stroke patients during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Farmers inspecting withered and dried-out crops due to the heat wave in Northeast Asia during July 2018"
] |
[
"People sweating and fanning themselves in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Temperature readings displayed on outdoor thermometers showing extremely high numbers during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queueing for water or other cooling supplies during the July 2018 heat wave in Japan, North Korea, South Korea, and China",
"Heat-related illnesses being treated at hospitals or medical tents during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Dried-up or wilted plants and crops due to the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia"
] |
[
"People fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queuing for water in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"thermometer screens displaying extremely high temperatures during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Trees and plants wilting due to the heat wave in Japan, North Korea, South Korea, China, and Northeast Asia in July 2018",
"People wearing masks and hats to protect themselves from the sun during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2018%EB%85%84_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%8F%AD%EC%97%BC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are commonly associated with extreme heat and its effects on people and the environment. These events should be visual and tangible, allowing them to be captured in a video, such as people trying to cool themselves, impacts on daily life, or consequences of the heat wave.\n\nEVENTS:\n1. People fanning themselves on the streets\n2. Citizens queuing for water in public spaces\n3. thermometer screens displaying extremely high temperatures\n4. Trees and plants wilting due to the heat\n5. People wearing masks and hats to protect themselves from the sun",
"en_article_excerpt": "Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.",
"en_article_url": "https://en.wikipedia.org/wiki/2018_Northeast_Asia_heat_wave",
"event_category": "disasters",
"event_name": "2018_northeast_asia_heat_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a heat wave, such as extreme weather conditions, people trying to cool themselves, and potential consequences of the heat wave. These events should be concrete and visual, allowing them to be easily depicted in a video.\n\nEVENTS:\n1. People sweating and fanning themselves on the streets\n2. Temperature readings on thermometers exceeding record highs\n3. Citizens gathering at public water parks and pools to cool off\n4. Ambulances rushing to treat heat stroke patients\n5. Farmers inspecting withered and dried-out crops due to the heat wave",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are visually representative of extreme heat and its effects on daily life and the environment. These events should be common consequences of a heat wave and plausible to be captured in a video during that time in the mentioned regions.\n\nEVENTS:\n1. People sweating and fanning themselves in public spaces\n2. Temperature readings displayed on outdoor thermometers showing extremely high numbers\n3. Citizens queueing for water or other cooling supplies\n4. Heat-related illnesses being treated at hospitals or medical tents\n5. Dried-up or wilted plants and crops due to the heat wave",
"video_URL": "http://youtube.com/watch?v=XYrtvA02lKc",
"video_description": "경기도 분당에서 폭염으로 수도 배관이 파열되면서 교량의 균열이 발생해 교통이 통제되고 있습니다. 어젯밤(29일) 10시쯤 경기도 성남시 분당구에 있는 길이 25m, 폭 20m인 야탑 10교에 설치된 수도 배관이 터지면서 물이 치솟았습니다. 이 사고로 교각이 왼쪽으로 15도 정도 기울었고 도로 일부에도 균열이 생겼습니다. 경찰은 사고가 나자 야탑교 사거리부터 탑골 사거리까지 440m 구간 차량 통행을 전면 통제했습니다. 사고 당시 다리 위를 지나던 차량이 많지 않아 별다른 피해는 없었습니다. 경기 분당경찰서는 노후화된 교각이 폭염 영향으로 팽창하면서 짓눌린 수도관이 터져 물이 치솟은 것으로 추정된다고 설명했습니다. 최기성 [choiks7@ytn.co.kr] ▶ 기사 원문 : http://www.ytn.co.kr/_ln/0103_201807300822376551 ▶ 제보 안내 : http://goo.gl/gEvsAL, 모바일앱, 8585@ytn.co.kr, #2424 ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ [ 한국 뉴스 채널 와이티엔 / Korea News Channel YTN ]",
"video_id": "XYrtvA02lKc"
}
| 4
|
[
"The image shows a road with visible cracks, suggesting structural damage. A bridge with a metal railing runs alongside the road. There are orange traffic cones and yellow caution tape blocking the area, indicating a restricted zone. The text in Korean suggests a severe issue with a 10-meter collapse and urges caution.",
"The image continues to depict a road with significant cracks, indicating ongoing structural damage. The bridge with a metal railing remains visible alongside the road. The area remains cordoned off with orange traffic cones and yellow caution tape, reinforcing the restricted zone status. The Korean text reiterates the severity of the 10-meter collapse and the need for caution, highlighting the ongoing nature of the problem.",
"The current frame shows a close-up view of a damaged concrete bridge structure. The bridge has a large, noticeable crack running along its length. There are fallen debris and a damaged section of the bridge visible, with a clear indication of structural failure. The area around the bridge is cordoned off with caution tape, and a person is partially visible near the bridge, suggesting a presence for inspection or maintenance. The Korean text emphasizes the ongoing issue with the bridge's structural integrity and the need for caution.",
"The current frame shows a different angle of the damaged concrete bridge structure. The bridge still has a large, noticeable crack running along its length, and more debris is visible on the ground below, indicating further structural failure. The area around the bridge remains cordoned off, emphasizing the hazardous situation. A person is visible in the background near the bridge, continuing the inspection or maintenance process. The Korean text reinforces the ongoing issue with the bridge's structural integrity and the need for caution, urging people to keep away from the damaged area."
] |
XYrtvA02lKc
|
A video depicts a severely damaged road and bridge structure, with visible cracks and a 10-meter collapse. The area is restricted, marked by orange traffic cones and yellow caution tape, indicating a hazardous situation. The footage shows a close-up of the bridge's structural failure, with fallen debris and a large crack along its length. A person, likely an inspector or maintenance worker, is present near the bridge, assessing the damage. The video highlights the ongoing issue with the bridge's structural integrity, emphasizing the need for caution and urging people to keep away from the damaged area.
|
Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.
|
[
"People sweating and fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Temperature readings on thermometers exceeding record highs during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens gathering at public water parks and pools to cool off during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Ambulances rushing to treat heat stroke patients during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Farmers inspecting withered and dried-out crops due to the heat wave in Northeast Asia during July 2018"
] |
[
"People sweating and fanning themselves in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Temperature readings displayed on outdoor thermometers showing extremely high numbers during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queueing for water or other cooling supplies during the July 2018 heat wave in Japan, North Korea, South Korea, and China",
"Heat-related illnesses being treated at hospitals or medical tents during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Dried-up or wilted plants and crops due to the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia"
] |
[
"People fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queuing for water in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"thermometer screens displaying extremely high temperatures during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Trees and plants wilting due to the heat wave in Japan, North Korea, South Korea, China, and Northeast Asia in July 2018",
"People wearing masks and hats to protect themselves from the sun during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2018%EB%85%84_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%8F%AD%EC%97%BC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are commonly associated with extreme heat and its effects on people and the environment. These events should be visual and tangible, allowing them to be captured in a video, such as people trying to cool themselves, impacts on daily life, or consequences of the heat wave.\n\nEVENTS:\n1. People fanning themselves on the streets\n2. Citizens queuing for water in public spaces\n3. thermometer screens displaying extremely high temperatures\n4. Trees and plants wilting due to the heat\n5. People wearing masks and hats to protect themselves from the sun",
"en_article_excerpt": "Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.",
"en_article_url": "https://en.wikipedia.org/wiki/2018_Northeast_Asia_heat_wave",
"event_category": "disasters",
"event_name": "2018_northeast_asia_heat_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a heat wave, such as extreme weather conditions, people trying to cool themselves, and potential consequences of the heat wave. These events should be concrete and visual, allowing them to be easily depicted in a video.\n\nEVENTS:\n1. People sweating and fanning themselves on the streets\n2. Temperature readings on thermometers exceeding record highs\n3. Citizens gathering at public water parks and pools to cool off\n4. Ambulances rushing to treat heat stroke patients\n5. Farmers inspecting withered and dried-out crops due to the heat wave",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are visually representative of extreme heat and its effects on daily life and the environment. These events should be common consequences of a heat wave and plausible to be captured in a video during that time in the mentioned regions.\n\nEVENTS:\n1. People sweating and fanning themselves in public spaces\n2. Temperature readings displayed on outdoor thermometers showing extremely high numbers\n3. Citizens queueing for water or other cooling supplies\n4. Heat-related illnesses being treated at hospitals or medical tents\n5. Dried-up or wilted plants and crops due to the heat wave",
"video_URL": "http://twitter.com/i/status/1038669934375493632",
"video_description": "재난수준 폭염에도 올 여름 전력부족 없었다 올여름 재난 수준의 폭염으로 냉방 전력수요가 급증했지만 전력이 부족한 날은 없었음 7월과 8월 두 달 동안 공급예비율은 7일간을 제외, 모두 두 자릿수 이상을 유지 전력부족도 없었고, 전기세도 감면해준 문재인 정부",
"video_id": "1038669934375493632"
}
| 4
|
[
"The image shows three individuals wearing red vests with text on the back, standing in front of a large digital display screen. They are engaged in a discussion or analysis of the data shown on the screen, which includes various graphs and numerical data. The screen displays information in Korean, with one person pointing at the screen, possibly explaining or highlighting specific details. The text at the bottom of the image indicates a news broadcast or report, mentioning \"재난수준 폭염에도 올 여름 전력부족 없었다\" which translates to \"Despite the disaster-level heatwave this summer, there was no power shortage.\" The context suggests they are monitoring or discussing energy consumption or power supply during a heatwave.",
"The image shows a close-up of a digital display screen with various numerical data and text in Korean, indicating a real-time update or report. A person is seen from the side, focusing intently on the screen, possibly analyzing the data displayed. The text at the bottom of the image mentions \"재난수준 폭염에도 올 여름 전력부족 없었다\" which translates to \"Despite the disaster-level heatwave this summer, there was no power shortage.\" The context suggests ongoing monitoring or discussion of energy consumption or power supply during the heatwave, consistent with the previous frame.",
"In the current frame, a person wearing a yellow vest with red text, possibly indicating a role in operations or management, is seen from behind facing a complex array of control panels. The individual is likely monitoring or adjusting settings on the panels, which are filled with numerous buttons and indicators. A computer screen is visible on the right, displaying multiple camera feeds and data, suggesting real-time surveillance or system monitoring. The text at the bottom of the screen reiterates the message \"재난수준 폭염에도 올 여름 전력부족 없었다,\" confirming the absence of power shortages during the heatwave. The person's focused posture and the detailed control panels indicate a high-stakes environment, possibly related to energy management or crisis response.",
"The scene continues to show the individual in the yellow vest with red text, maintaining focus on the control panels. The person's posture suggests ongoing monitoring or adjusting of the control settings, critical for managing the operations depicted. The computer screen still displays multiple camera feeds and data, essential for real-time surveillance or system monitoring. The text reiterates the absence of power shortages during the heatwave, emphasizing the successful management of the situation. The environment remains high-stakes, highlighting the importance of the individual's role in ensuring system stability and crisis response."
] |
1038669934375493632
|
A video depicting a team of individuals monitoring and managing energy consumption or power supply during a disaster-level heatwave. The team, wearing red and yellow vests, is shown analyzing data on digital display screens, discussing the information, and adjusting control panels to ensure a stable power supply. The context suggests a high-stakes environment, possibly an energy management or crisis response center, where the team is working to prevent power shortages despite the extreme weather conditions. The video emphasizes the team's successful management of the situation, as indicated by the repeated message "Despite the disaster-level heatwave this summer, there was no power shortage." The overall event appears to be a news broadcast or report highlighting the team's efforts and the effective management of the power supply during the heatwave.
|
Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.
|
[
"People sweating and fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Temperature readings on thermometers exceeding record highs during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens gathering at public water parks and pools to cool off during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Ambulances rushing to treat heat stroke patients during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Farmers inspecting withered and dried-out crops due to the heat wave in Northeast Asia during July 2018"
] |
[
"People sweating and fanning themselves in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Temperature readings displayed on outdoor thermometers showing extremely high numbers during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queueing for water or other cooling supplies during the July 2018 heat wave in Japan, North Korea, South Korea, and China",
"Heat-related illnesses being treated at hospitals or medical tents during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Dried-up or wilted plants and crops due to the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia"
] |
[
"People fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queuing for water in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"thermometer screens displaying extremely high temperatures during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Trees and plants wilting due to the heat wave in Japan, North Korea, South Korea, China, and Northeast Asia in July 2018",
"People wearing masks and hats to protect themselves from the sun during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2018%EB%85%84_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%8F%AD%EC%97%BC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are commonly associated with extreme heat and its effects on people and the environment. These events should be visual and tangible, allowing them to be captured in a video, such as people trying to cool themselves, impacts on daily life, or consequences of the heat wave.\n\nEVENTS:\n1. People fanning themselves on the streets\n2. Citizens queuing for water in public spaces\n3. thermometer screens displaying extremely high temperatures\n4. Trees and plants wilting due to the heat\n5. People wearing masks and hats to protect themselves from the sun",
"en_article_excerpt": "Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.",
"en_article_url": "https://en.wikipedia.org/wiki/2018_Northeast_Asia_heat_wave",
"event_category": "disasters",
"event_name": "2018_northeast_asia_heat_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a heat wave, such as extreme weather conditions, people trying to cool themselves, and potential consequences of the heat wave. These events should be concrete and visual, allowing them to be easily depicted in a video.\n\nEVENTS:\n1. People sweating and fanning themselves on the streets\n2. Temperature readings on thermometers exceeding record highs\n3. Citizens gathering at public water parks and pools to cool off\n4. Ambulances rushing to treat heat stroke patients\n5. Farmers inspecting withered and dried-out crops due to the heat wave",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are visually representative of extreme heat and its effects on daily life and the environment. These events should be common consequences of a heat wave and plausible to be captured in a video during that time in the mentioned regions.\n\nEVENTS:\n1. People sweating and fanning themselves in public spaces\n2. Temperature readings displayed on outdoor thermometers showing extremely high numbers\n3. Citizens queueing for water or other cooling supplies\n4. Heat-related illnesses being treated at hospitals or medical tents\n5. Dried-up or wilted plants and crops due to the heat wave",
"video_URL": "http://twitter.com/i/status/1026325665824403456",
"video_description": "\"너무 덥지? 잘 지내고 있어?\" 관심어린 안부 한 마디가 폭염의 열기를 식힐 수 있습니다. 더위가 기승을 부리는 요즘 부모님께 또 친구에게 안부 전화 하는 건 어떨까요? (출처 : 국민안전방송 안전한TV) #성남시 #폭염 #가마솥더위 #폭염대비",
"video_id": "1026325665824403456"
}
| 4
|
[
"A woman is seated, appearing to be engaged in a serious conversation or interview. She has a focused expression on her face, suggesting attentiveness. The background is blurred, featuring bookshelves, indicating a setting that might be an office or study room.",
"The woman maintains her serious expression, continuing the conversation or interview. The setting remains consistent, with bookshelves in the blurred background, reinforcing the professional or academic environment.",
"The woman, now seen from the back, appears to be on a phone call, standing on what looks like a balcony or an elevated area with a cityscape in the background. She is wearing a light-colored blouse and has her hair tied back. Text on the screen, written in Korean, reads \"폭염엔 부모님의 안부를 물으세요\" which translates to \"During the heatwave, ask about your parents' well-being.\" The logo for \"안전한\" TV is visible in the top right corner, indicating a public service announcement or a segment on a television program emphasizing safety and checking on family during extreme weather conditions.",
"The woman remains on the balcony or elevated area, still engaged in a phone call. She maintains her position with the cityscape in the background. The text \"폭염엔 부모님의 안부를 물으세요\" continues to be displayed, reinforcing the message to check on parents during a heatwave. The \"안전한\" TV logo is still present, suggesting the ongoing public service announcement theme focused on safety and familial well-being in extreme weather conditions."
] |
1026325665824403456
|
A woman is featured in a video that appears to be a public service announcement, emphasizing the importance of checking on family members, specifically parents, during extreme weather conditions such as heatwaves. The video transitions from an indoor setting, possibly an office or study room where she is engaged in a serious conversation, to an outdoor location with a cityscape view, where she makes a phone call. Throughout the video, the message "During the heatwave, ask about your parents' well-being" is displayed, along with the logo of "안전한" TV, indicating a focus on safety and familial well-being. The woman's actions and the context suggest a narrative encouraging viewers to prioritize their family's safety and well-being, particularly during adverse weather conditions.
|
Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.
|
[
"People sweating and fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Temperature readings on thermometers exceeding record highs during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens gathering at public water parks and pools to cool off during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Ambulances rushing to treat heat stroke patients during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Farmers inspecting withered and dried-out crops due to the heat wave in Northeast Asia during July 2018"
] |
[
"People sweating and fanning themselves in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Temperature readings displayed on outdoor thermometers showing extremely high numbers during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queueing for water or other cooling supplies during the July 2018 heat wave in Japan, North Korea, South Korea, and China",
"Heat-related illnesses being treated at hospitals or medical tents during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Dried-up or wilted plants and crops due to the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia"
] |
[
"People fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queuing for water in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"thermometer screens displaying extremely high temperatures during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Trees and plants wilting due to the heat wave in Japan, North Korea, South Korea, China, and Northeast Asia in July 2018",
"People wearing masks and hats to protect themselves from the sun during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2018%EB%85%84_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%8F%AD%EC%97%BC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are commonly associated with extreme heat and its effects on people and the environment. These events should be visual and tangible, allowing them to be captured in a video, such as people trying to cool themselves, impacts on daily life, or consequences of the heat wave.\n\nEVENTS:\n1. People fanning themselves on the streets\n2. Citizens queuing for water in public spaces\n3. thermometer screens displaying extremely high temperatures\n4. Trees and plants wilting due to the heat\n5. People wearing masks and hats to protect themselves from the sun",
"en_article_excerpt": "Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.",
"en_article_url": "https://en.wikipedia.org/wiki/2018_Northeast_Asia_heat_wave",
"event_category": "disasters",
"event_name": "2018_northeast_asia_heat_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a heat wave, such as extreme weather conditions, people trying to cool themselves, and potential consequences of the heat wave. These events should be concrete and visual, allowing them to be easily depicted in a video.\n\nEVENTS:\n1. People sweating and fanning themselves on the streets\n2. Temperature readings on thermometers exceeding record highs\n3. Citizens gathering at public water parks and pools to cool off\n4. Ambulances rushing to treat heat stroke patients\n5. Farmers inspecting withered and dried-out crops due to the heat wave",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are visually representative of extreme heat and its effects on daily life and the environment. These events should be common consequences of a heat wave and plausible to be captured in a video during that time in the mentioned regions.\n\nEVENTS:\n1. People sweating and fanning themselves in public spaces\n2. Temperature readings displayed on outdoor thermometers showing extremely high numbers\n3. Citizens queueing for water or other cooling supplies\n4. Heat-related illnesses being treated at hospitals or medical tents\n5. Dried-up or wilted plants and crops due to the heat wave",
"video_URL": "http://youtube.com/watch?v=09W_DRDb9HQ",
"video_description": "+--------------------------------------------------------------------------------+ ◈ 중앙일보 공식 SNS 채널 ◈ - 페이스북 : https://www.facebook.com/joongang/ - 인스타그램 : https://www.instagram.com/joongangilbo - 트위터 : http://www.twitter.com/joongangilbo - 유튜브 : https://goo.gl/DyGoPC - 네이버 TV : http://tvcast.naver.com/joongangsisa - 네이버포스트 : https://goo.gl/eeYh4k 당신이 원하는 뉴스, 중앙일보 SNS 채널과 함께하세요. +--------------------------------------------------------------------------------+",
"video_id": "09W_DRDb9HQ"
}
| 4
|
[
"The image shows a frying pan with a piece of cheese, an egg with a runny yolk, and several slices of bacon. There is a text overlay in Korean, along with the time \"1 hr 48 min\" on a timer icon.",
"The current frame shows a close-up of a piece of cheese with a portion scooped out, using a wooden spoon. The cheese appears to be a soft, white variety, possibly Brie or Camembert. The focus is on the texture and consistency of the cheese, which looks creamy and spreadable. The background is minimalistic, emphasizing the cheese. The text overlay in Korean is visible at the top left corner, and the scene appears to be part of a culinary demonstration or food review, possibly continuing the theme of showcasing different types of cheese or food preparation techniques.",
"The current frame depicts a transparent plastic cup against a blue sky background, partially filled with a dark red liquid, which appears to be blood. The focus is on the blood's consistency and texture, with some splashes visible at the top. A timer overlay indicates \"10 min,\" suggesting a time lapse or demonstration is taking place. The text overlay in Korean reads \"대한민국이 될 15분 만에 녹아버린 초콜릿,\" which translates to \"15 minutes until Korea melts chocolate.\" This frame continues the culinary demonstration or food review theme, possibly illustrating a time-dependent transformation of the substance, likely related to heat or another environmental factor.",
"The current frame shows a dark screen with white Korean text that reads \"종영일보,\" which translates to \"Closing day report.\" This indicates a summary or conclusion segment of the video, possibly reviewing or summarizing the content or events discussed or demonstrated throughout the video, such as the melting chocolate demonstration mentioned previously."
] |
09W_DRDb9HQ
|
A culinary video showcases various food items and demonstrations, including a frying pan with cheese, egg, and bacon, and a close-up of soft, white cheese being scooped with a wooden spoon. The video also features an unusual scene with a cup of blood against a blue sky background, possibly as part of a time-lapse demonstration. Later, a segment appears to involve melting chocolate, with a timer counting down. The video concludes with a summary or closing segment, titled "Closing day report," likely reviewing the culinary content and demonstrations presented throughout the video.
|
Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.
|
[
"People sweating and fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Temperature readings on thermometers exceeding record highs during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens gathering at public water parks and pools to cool off during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Ambulances rushing to treat heat stroke patients during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Farmers inspecting withered and dried-out crops due to the heat wave in Northeast Asia during July 2018"
] |
[
"People sweating and fanning themselves in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Temperature readings displayed on outdoor thermometers showing extremely high numbers during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queueing for water or other cooling supplies during the July 2018 heat wave in Japan, North Korea, South Korea, and China",
"Heat-related illnesses being treated at hospitals or medical tents during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Dried-up or wilted plants and crops due to the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia"
] |
[
"People fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queuing for water in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"thermometer screens displaying extremely high temperatures during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Trees and plants wilting due to the heat wave in Japan, North Korea, South Korea, China, and Northeast Asia in July 2018",
"People wearing masks and hats to protect themselves from the sun during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2018%EB%85%84_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%8F%AD%EC%97%BC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are commonly associated with extreme heat and its effects on people and the environment. These events should be visual and tangible, allowing them to be captured in a video, such as people trying to cool themselves, impacts on daily life, or consequences of the heat wave.\n\nEVENTS:\n1. People fanning themselves on the streets\n2. Citizens queuing for water in public spaces\n3. thermometer screens displaying extremely high temperatures\n4. Trees and plants wilting due to the heat\n5. People wearing masks and hats to protect themselves from the sun",
"en_article_excerpt": "Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.",
"en_article_url": "https://en.wikipedia.org/wiki/2018_Northeast_Asia_heat_wave",
"event_category": "disasters",
"event_name": "2018_northeast_asia_heat_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a heat wave, such as extreme weather conditions, people trying to cool themselves, and potential consequences of the heat wave. These events should be concrete and visual, allowing them to be easily depicted in a video.\n\nEVENTS:\n1. People sweating and fanning themselves on the streets\n2. Temperature readings on thermometers exceeding record highs\n3. Citizens gathering at public water parks and pools to cool off\n4. Ambulances rushing to treat heat stroke patients\n5. Farmers inspecting withered and dried-out crops due to the heat wave",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are visually representative of extreme heat and its effects on daily life and the environment. These events should be common consequences of a heat wave and plausible to be captured in a video during that time in the mentioned regions.\n\nEVENTS:\n1. People sweating and fanning themselves in public spaces\n2. Temperature readings displayed on outdoor thermometers showing extremely high numbers\n3. Citizens queueing for water or other cooling supplies\n4. Heat-related illnesses being treated at hospitals or medical tents\n5. Dried-up or wilted plants and crops due to the heat wave",
"video_URL": "http://youtube.com/watch?v=8wcKX0LCc08",
"video_description": "폭염 때문에 힘든 건 사람만이 아닙니다. 과육이 물러진 수박이 산지에서 폐기되는 등 제철 과일과 채소 피해도 잇따르고 있습니다.",
"video_id": "8wcKX0LCc08"
}
| 4
|
[
"The video displays a field with numerous watermelons scattered on the ground. Some of the watermelons are intact while others are cut open, revealing the red flesh inside. The setting appears to be a farm or a garden with green foliage surrounding the watermelons.",
"In the field, a man is kneeling on the ground next to the watermelons, appearing to be speaking or giving a report, possibly about the watermelons or the situation in the field.",
"The current frame is a close-up shot of a corn plant's tassel, which appears dry and withered, suggesting a lack of moisture. The focus is on the details of the plant, highlighting the dry conditions. There is no visible human activity in this frame, but the context from the previous frame suggests the scene is related to agricultural issues, potentially due to drought, given the dry appearance of the corn. The time \"07:19\" and the weather indicator \"대구 26°\" (Daegu 26 degrees) provide additional context, indicating the time and location of the video.",
"A news anchor is presenting a segment on a news program, indicated by the \"NEWS TODAY\" logo and the time \"07:19\" displayed on the screen. The anchor is dressed in an orange suit and is holding papers, likely containing the news content. The television screen in the background shows the title \"뉴스투데이\" (News Today) and the word \"오늘의 날씨\" (Today's Weather), suggesting that the segment might be discussing the weather or related news. The lower third of the screen displays the temperature in Seoul as 30 degrees, reinforcing the weather-related topic."
] |
8wcKX0LCc08
|
A news segment is being presented, focusing on agricultural issues, likely related to a drought. The report begins with footage of a field containing numerous watermelons, some of which are cut open, and a man kneeling next to them, possibly discussing the condition of the crops. The scene then cuts to a close-up of a withered corn plant, emphasizing the dry conditions. The news anchor is then shown presenting the segment, with the background screen displaying information about the current weather, including the temperature in Daegu and Seoul, highlighting the relevance of the weather to the agricultural issues being discussed.
|
Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.
|
[
"People sweating and fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Temperature readings on thermometers exceeding record highs during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens gathering at public water parks and pools to cool off during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"Ambulances rushing to treat heat stroke patients during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Farmers inspecting withered and dried-out crops due to the heat wave in Northeast Asia during July 2018"
] |
[
"People sweating and fanning themselves in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Temperature readings displayed on outdoor thermometers showing extremely high numbers during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queueing for water or other cooling supplies during the July 2018 heat wave in Japan, North Korea, South Korea, and China",
"Heat-related illnesses being treated at hospitals or medical tents during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Dried-up or wilted plants and crops due to the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia"
] |
[
"People fanning themselves on the streets during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Citizens queuing for water in public spaces during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia",
"thermometer screens displaying extremely high temperatures during the July 2018 heat wave in Japan, North Korea, South Korea, China, and Northeast Asia",
"Trees and plants wilting due to the heat wave in Japan, North Korea, South Korea, China, and Northeast Asia in July 2018",
"People wearing masks and hats to protect themselves from the sun during the July 2018 heat wave in Japan, North Korea, South Korea, China, and other parts of Northeast Asia"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2018%EB%85%84_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%8F%AD%EC%97%BC",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are commonly associated with extreme heat and its effects on people and the environment. These events should be visual and tangible, allowing them to be captured in a video, such as people trying to cool themselves, impacts on daily life, or consequences of the heat wave.\n\nEVENTS:\n1. People fanning themselves on the streets\n2. Citizens queuing for water in public spaces\n3. thermometer screens displaying extremely high temperatures\n4. Trees and plants wilting due to the heat\n5. People wearing masks and hats to protect themselves from the sun",
"en_article_excerpt": "Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.",
"en_article_url": "https://en.wikipedia.org/wiki/2018_Northeast_Asia_heat_wave",
"event_category": "disasters",
"event_name": "2018_northeast_asia_heat_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a heat wave, such as extreme weather conditions, people trying to cool themselves, and potential consequences of the heat wave. These events should be concrete and visual, allowing them to be easily depicted in a video.\n\nEVENTS:\n1. People sweating and fanning themselves on the streets\n2. Temperature readings on thermometers exceeding record highs\n3. Citizens gathering at public water parks and pools to cool off\n4. Ambulances rushing to treat heat stroke patients\n5. Farmers inspecting withered and dried-out crops due to the heat wave",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Throughout much of July 2018, a record-breaking heat wave affected large areas of Northeast Asia including Japan, North Korea, South Korea, China.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about a record-breaking heat wave in Northeast Asia in July 2018, it makes sense to extract events that are visually representative of extreme heat and its effects on daily life and the environment. These events should be common consequences of a heat wave and plausible to be captured in a video during that time in the mentioned regions.\n\nEVENTS:\n1. People sweating and fanning themselves in public spaces\n2. Temperature readings displayed on outdoor thermometers showing extremely high numbers\n3. Citizens queueing for water or other cooling supplies\n4. Heat-related illnesses being treated at hospitals or medical tents\n5. Dried-up or wilted plants and crops due to the heat wave",
"video_URL": "http://youtube.com/watch?v=g9CbgCp52qc",
"video_description": "연일 불볕더위가 기승을 부리는 가운데 서울시가 도로에 물을 뿌려 열을 식히는 작업을 강화합니다. 서울시는 낮 최고기온이 32℃ 이상으로 오르면 폭염에 노출될 가능성이 큰 정오부터 오후 4시까지 중앙버스전용차로를 중심으로 살수 차량 운행을 확대한다고 밝혔습니다. 서울시는 여름철 도로 물뿌리기로 미세먼지를 제거하는 동시에 온열질환 등 폭염으로 발생할 수 있는 안전사고도 방지할 수 있기를 기대하고 있습니다. ▶ 기사 원문 : http://www.ytn.co.kr/_ln/0115_201807181131189139 ▶ 제보 안내 : http://goo.gl/gEvsAL, 모바일앱, 8585@ytn.co.kr, #2424 ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ [ 한국 뉴스 채널 와이티엔 / Korea News Channel YTN ]",
"video_id": "g9CbgCp52qc"
}
| 4
|
[
"The current frame displays a dark screen with partially visible text that reads \"LIC ISG.\"",
"A man with a serious expression on his face is captured in a close-up shot. He has long dark hair and a beard, and he appears to be contemplating or worried about something. The background is out of focus but seems to be an outdoor setting with lights, suggesting it might be evening or nighttime.",
"A man wearing a black beanie and a heavy coat stands in a dimly lit setting, possibly indoors, with a serious expression on his face. The background includes a metallic door and a lamp, adding to the somber atmosphere of the scene.",
"A person with a black beanie and a heavy coat is running in a dimly lit corridor, possibly in a hurry or fleeing from something. The environment remains consistent with a metallic door and lamp in the background, maintaining the somber atmosphere."
] |
g9CbgCp52qc
|
The video appears to depict a serious and potentially ominous scene involving a man. It begins with a dark screen displaying partial text "LIC ISG," followed by a close-up shot of the man with a worried expression, set against an outdoor evening or nighttime backdrop. The scene then shifts to the man, now wearing a black beanie and heavy coat, standing in a dimly lit indoor setting with a serious expression. The atmosphere remains somber as the man is later shown running down a corridor, possibly in a hurry or fleeing, maintaining the indoor setting with metallic doors and lamps. The overall context suggests a narrative of concern, urgency, or danger, with the man as the central figure navigating through these ominous environments.
|
In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.
|
[
"People bundling up in heavy clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in heavy clothing to brave the cold during record cold temperatures in East Asia, Southeast Asia, and South Asia in late January 2016",
"People bundling up in heavy clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong, during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia during late January 2016 snowfall events",
"Frost or ice formation on car windshields, roads, and buildings during the cold wave in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings during snowfall in East Asia, Southeast Asia, and South Asia in late January 2016",
"Residents in East Asia, Southeast Asia, and South Asia shoveling snow or using umbrellas to navigate snowy streets during the late January 2016 cold wave",
"Residents in East and South Asia shoveling snow or using umbrellas to navigate snowy streets during record cold temperatures in late January 2016",
"Residents shoveling snow or using umbrellas to navigate snowy streets in East Asia, Southeast Asia, or South Asia in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia Southeast Asia and South Asia during late January 2016"
] |
[
"People bundling up in warm clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in warm clothing to brave the cold in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People bundling up in warm clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"Frozen lakes and rivers with people ice skating or playing in East Asia Southeast Asia or South Asia during late January 2016 cold wave",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia where people are ice skating or playing during the record cold temperatures of late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, or South Asia where people are ice skating or playing in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia during late January 2016 cold wave",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia after snowfall in late January 2016",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 cold wave",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off sidewalks and driveways in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
[
"People bundling up in winter clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in winter clothing to brave the cold in East and South Asia during late January 2016 record cold snap",
"People bundling up in winter clothing to brave the cold during late January 2016 snowfall in East, Southeast, and South Asia",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during the cold wave of late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia due to record cold temperatures in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia after snowfall in late January 2016",
"Trees and buildings covered in snow and ice during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off streets and sidewalks in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_1%EC%9B%94_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%95%9C%ED%8C%8C",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, and the consequences of the snowfall, as these events are likely to be visualized in a video. The extracted events should be specific to the regions mentioned and the time of year, taking into account the typical climate and scenery of East Asia, mainland Southeast Asia, and northern South Asia in late January.\n\nEVENTS:\n1. People bundling up in winter clothing to brave the cold\n2. Snow falling in cities and towns\n3. Frozen lakes and rivers\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off streets and sidewalks",
"en_article_excerpt": "In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.",
"en_article_url": "https://en.wikipedia.org/wiki/January_2016_East_Asia_cold_wave",
"event_category": "disasters",
"event_name": "2016_east_asia_cold_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, disruptions to daily life, and unusual weather phenomena like snowfall in regions that rarely see it. These events would be visually representative of the cold wave's impact and could be reasonably captured in a video.\n\nEVENTS:\n1. People bundling up in heavy clothing to brave the cold\n2. Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong\n3. Frost or ice forming on surfaces like car windshields, roads, and buildings\n4. Residents shoveling snow or using umbrellas to navigate snowy streets\n5. Cars and buses driving slowly on icy or snow-covered roads",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a cold wave, such as people dealing with the cold and snow, and disruptions to daily life. These events should be visual and concrete, and could be reasonably captured in a video of the cold wave's impact.\n\nEVENTS:\n1. People bundling up in warm clothing to brave the cold\n2. Snowplows clearing roads and highways\n3. Frozen lakes and rivers with people ice skating or playing\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off sidewalks and driveways",
"video_URL": "http://twitter.com/i/status/1346958705632296961",
"video_description": "전국 눈 대단한 한파 다들 잘 이겨내요",
"video_id": "1346958705632296961"
}
| 4
|
[
"The frame depicts a residential area with several tall apartment buildings. The buildings have a green and white color scheme and are arranged in a linear fashion along a pathway that leads through a landscaped area with trees and pathways. The sky appears overcast, and the scene is relatively quiet with no visible people or movement.",
"The frame depicts a similar view of the residential area, still featuring the tall apartment buildings with a green and white color scheme. A person is now visible walking along the pathway, moving from left to right across the frame. The sky remains overcast, and the overall scene is quiet. The presence of the person walking adds a slight dynamic element to the previously still setting.",
"The current frame continues to show the residential area with tall apartment buildings. The person who was previously seen walking along the pathway has progressed further, moving more towards the right of the frame, maintaining the slight dynamic element in the otherwise static scene. The overcast sky and the quiet atmosphere persist, with no significant changes in the environment or additional activities.",
"The current frame continues to depict the residential area with tall apartment buildings. The person who was previously seen walking along the pathway is no longer visible, suggesting they have moved out of the frame or paused their movement. The overcast sky and the quiet atmosphere persist, with no significant changes in the environment or additional activities. The scene remains largely static with no new dynamic elements introduced."
] |
1346958705632296961
|
The video depicts a quiet scene in a residential area with tall apartment buildings, set against an overcast sky. A person is seen walking along a pathway through a landscaped area, moving from left to right across the frame, introducing a brief dynamic element into the otherwise static scene. The person's movement is the primary action, and they eventually move out of view, leaving the scene to return to its initial quiet and still state. The setting remains unchanged throughout, with no other notable activities or participants.
|
In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.
|
[
"People bundling up in heavy clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in heavy clothing to brave the cold during record cold temperatures in East Asia, Southeast Asia, and South Asia in late January 2016",
"People bundling up in heavy clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong, during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia during late January 2016 snowfall events",
"Frost or ice formation on car windshields, roads, and buildings during the cold wave in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings during snowfall in East Asia, Southeast Asia, and South Asia in late January 2016",
"Residents in East Asia, Southeast Asia, and South Asia shoveling snow or using umbrellas to navigate snowy streets during the late January 2016 cold wave",
"Residents in East and South Asia shoveling snow or using umbrellas to navigate snowy streets during record cold temperatures in late January 2016",
"Residents shoveling snow or using umbrellas to navigate snowy streets in East Asia, Southeast Asia, or South Asia in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia Southeast Asia and South Asia during late January 2016"
] |
[
"People bundling up in warm clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in warm clothing to brave the cold in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People bundling up in warm clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"Frozen lakes and rivers with people ice skating or playing in East Asia Southeast Asia or South Asia during late January 2016 cold wave",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia where people are ice skating or playing during the record cold temperatures of late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, or South Asia where people are ice skating or playing in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia during late January 2016 cold wave",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia after snowfall in late January 2016",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 cold wave",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off sidewalks and driveways in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
[
"People bundling up in winter clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in winter clothing to brave the cold in East and South Asia during late January 2016 record cold snap",
"People bundling up in winter clothing to brave the cold during late January 2016 snowfall in East, Southeast, and South Asia",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during the cold wave of late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia due to record cold temperatures in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia after snowfall in late January 2016",
"Trees and buildings covered in snow and ice during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off streets and sidewalks in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_1%EC%9B%94_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%95%9C%ED%8C%8C",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, and the consequences of the snowfall, as these events are likely to be visualized in a video. The extracted events should be specific to the regions mentioned and the time of year, taking into account the typical climate and scenery of East Asia, mainland Southeast Asia, and northern South Asia in late January.\n\nEVENTS:\n1. People bundling up in winter clothing to brave the cold\n2. Snow falling in cities and towns\n3. Frozen lakes and rivers\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off streets and sidewalks",
"en_article_excerpt": "In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.",
"en_article_url": "https://en.wikipedia.org/wiki/January_2016_East_Asia_cold_wave",
"event_category": "disasters",
"event_name": "2016_east_asia_cold_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, disruptions to daily life, and unusual weather phenomena like snowfall in regions that rarely see it. These events would be visually representative of the cold wave's impact and could be reasonably captured in a video.\n\nEVENTS:\n1. People bundling up in heavy clothing to brave the cold\n2. Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong\n3. Frost or ice forming on surfaces like car windshields, roads, and buildings\n4. Residents shoveling snow or using umbrellas to navigate snowy streets\n5. Cars and buses driving slowly on icy or snow-covered roads",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a cold wave, such as people dealing with the cold and snow, and disruptions to daily life. These events should be visual and concrete, and could be reasonably captured in a video of the cold wave's impact.\n\nEVENTS:\n1. People bundling up in warm clothing to brave the cold\n2. Snowplows clearing roads and highways\n3. Frozen lakes and rivers with people ice skating or playing\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off sidewalks and driveways",
"video_URL": "http://youtube.com/watch?v=2qCY0mBeZy4",
"video_description": "지난 연말부터 시작된 한파의 기세는 여전했습니다. 강원도 산간은 영하 20도, 철원 영하 15.9도, 서울도 영하 9.8도를 기록했습니다. 이번 한파는 풀리지 않고 다음 주까지 이어질 것으로 보입니다. 주말부터 다음 주 초반까지 기온이 조금 오르지만, 이후 다시 기온이 급강하하기 때문입니다. 특히 다음 주 후반에는 서울 기온이 영하 15도까지 떨어지는 강력한 한파가 맹위를 떨칠 전망입니다. 올겨울 들어 가장 강력한 추위입니다. [윤기한 / 기상청 통보관 : 한파가 누그러지지 않은 상황에서 2차로 한기가 추가로 내려오는 것입니다. 따라서 다음 주에는 추위 강도가 지금보다 훨씬 세질 것으로 보입니다.] 북극 한파가 새해 첫날부터 열흘 이상 이어지는 것입니다. 서해안의 눈도 잠시 약해졌다가 다음 주 중반 이후에는 눈 폭탄으로 쏟아질 것으로 우려됩니다. 기상청은 이번 한파와 폭설로 동파와 축사, 비닐하우스 붕괴 등 시설물 피해가 속출할 것이라며 사전 대비를 당부했습니다. 취재기자: 김진두 영상편집: 한경희 자막뉴스: 박해진 ▶ 기사 원문 : https://www.ytn.co.kr/replay/view.php?idx=29&key=202101020712296278 ▶ 제보 안내 : http://goo.gl/gEvsAL, 모바일앱, social@ytn.co.kr, #2424 ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN & YTN plus 무단 전재 및 재배포금지",
"video_id": "2qCY0mBeZy4"
}
| 4
|
[
"The image shows people walking in a line on a bridge. They are wearing winter clothing and masks. Text in Korean is displayed on the screen, indicating the movement of the ground from the front of the bridge to the back.",
"The current frame shows a large blue container with a wooden stick being used to stir or mix something inside. A person wearing a red jacket and white gloves is operating a red device, possibly a heater, to heat the contents of the container. The snowy background suggests a cold environment. The text in Korean at the bottom reads \"올겨울 들어 가장 강력한 추위입니다.\" which translates to \"This is the strongest cold wave of this winter.\" The scene suggests that the people are taking precautions against the extreme cold.",
"The current frame shows a child in a blue jacket and hood, with snow falling heavily around them. They appear to be shivering, indicating the cold weather. In the background, another person in a blue jacket can be seen, suggesting a group activity. The text at the bottom reads \"서해안의 눈도 잠시 약해졌다가 다시 중반 이후에는 눈 폭탄으로 쏟아질 것으로 우려됩니다.\" which translates to \"The snow in the western coastal area weakened for a while but is expected to return to a snowstorm again in the second half.\" This indicates concern about the ongoing heavy snowfall and the need for continued precautions.",
"The video frames depict a snowy outdoor scene with a child wearing a blue jacket and hood, covered in snow, standing amidst heavy snowfall. The child appears cold, likely due to the significant snow accumulation. Another individual, also in a blue jacket, is present, indicating a shared experience in the snowy environment. Text on the image warns of a return to heavy snowfall in the western coastal area, emphasizing the need for precautions and highlighting the ongoing snowy conditions."
] |
2qCY0mBeZy4
|
The video depicts a scene of people dealing with a strong cold wave and heavy snowfall, likely in South Korea. The event shows individuals taking precautions against the extreme cold, such as wearing winter clothing and masks, and using heaters to warm up. A group of people, including children, are shown braving the harsh weather conditions, with snow falling heavily around them. The video highlights the need for continued precautions as the snowfall is expected to return and worsen in the western coastal area. Overall, the video captures the challenges faced by people during a severe winter storm, emphasizing the importance of preparedness and safety measures.
|
In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.
|
[
"People bundling up in heavy clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in heavy clothing to brave the cold during record cold temperatures in East Asia, Southeast Asia, and South Asia in late January 2016",
"People bundling up in heavy clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong, during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia during late January 2016 snowfall events",
"Frost or ice formation on car windshields, roads, and buildings during the cold wave in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings during snowfall in East Asia, Southeast Asia, and South Asia in late January 2016",
"Residents in East Asia, Southeast Asia, and South Asia shoveling snow or using umbrellas to navigate snowy streets during the late January 2016 cold wave",
"Residents in East and South Asia shoveling snow or using umbrellas to navigate snowy streets during record cold temperatures in late January 2016",
"Residents shoveling snow or using umbrellas to navigate snowy streets in East Asia, Southeast Asia, or South Asia in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia Southeast Asia and South Asia during late January 2016"
] |
[
"People bundling up in warm clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in warm clothing to brave the cold in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People bundling up in warm clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"Frozen lakes and rivers with people ice skating or playing in East Asia Southeast Asia or South Asia during late January 2016 cold wave",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia where people are ice skating or playing during the record cold temperatures of late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, or South Asia where people are ice skating or playing in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia during late January 2016 cold wave",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia after snowfall in late January 2016",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 cold wave",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off sidewalks and driveways in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
[
"People bundling up in winter clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in winter clothing to brave the cold in East and South Asia during late January 2016 record cold snap",
"People bundling up in winter clothing to brave the cold during late January 2016 snowfall in East, Southeast, and South Asia",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during the cold wave of late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia due to record cold temperatures in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia after snowfall in late January 2016",
"Trees and buildings covered in snow and ice during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off streets and sidewalks in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_1%EC%9B%94_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%95%9C%ED%8C%8C",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, and the consequences of the snowfall, as these events are likely to be visualized in a video. The extracted events should be specific to the regions mentioned and the time of year, taking into account the typical climate and scenery of East Asia, mainland Southeast Asia, and northern South Asia in late January.\n\nEVENTS:\n1. People bundling up in winter clothing to brave the cold\n2. Snow falling in cities and towns\n3. Frozen lakes and rivers\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off streets and sidewalks",
"en_article_excerpt": "In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.",
"en_article_url": "https://en.wikipedia.org/wiki/January_2016_East_Asia_cold_wave",
"event_category": "disasters",
"event_name": "2016_east_asia_cold_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, disruptions to daily life, and unusual weather phenomena like snowfall in regions that rarely see it. These events would be visually representative of the cold wave's impact and could be reasonably captured in a video.\n\nEVENTS:\n1. People bundling up in heavy clothing to brave the cold\n2. Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong\n3. Frost or ice forming on surfaces like car windshields, roads, and buildings\n4. Residents shoveling snow or using umbrellas to navigate snowy streets\n5. Cars and buses driving slowly on icy or snow-covered roads",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a cold wave, such as people dealing with the cold and snow, and disruptions to daily life. These events should be visual and concrete, and could be reasonably captured in a video of the cold wave's impact.\n\nEVENTS:\n1. People bundling up in warm clothing to brave the cold\n2. Snowplows clearing roads and highways\n3. Frozen lakes and rivers with people ice skating or playing\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off sidewalks and driveways",
"video_URL": "http://youtube.com/watch?v=QmU3W9ffZeM",
"video_description": "기온이 마치 롤러코스터처럼 떨어질 모양입니다. 얼마나 추워질지 벌써부터 겁도 나는데요. 일단 내일 아침 서울 기온은 영하 11도까지 떨어지겠고요. 목요일부터는 찬바람이 더욱 강해지면서 아침 기온이 영하 15도로 전날 낮 대비 13도가량 기온이 급감하겠습니다. 한낮 기온도 영하 11도에 머물 텐데요. 다음 날 금요일 아침 기온은 영하 17도로 이번 한파의 절정이 되겠습니다. 한파 대비 미리 해주시는 게 좋겠습니다. 이에 따라서 오늘 밤 9시를 기해서는 경기 동부와 북부, 충청도, 경북 곳곳에 한파경보가 발효되겠고요. 서울 등 그 밖의 수도권과 충청도 전북 내륙 지역까지 한파주의보가 확대 발효되겠습니다. 눈구름도 함께 몰려오겠습니다. 내일 퇴근길 무렵에 서해 등 서쪽지역에 눈이 내리기 시작해 밤사이에 전국으로 확대될 텐데요. 특히 서해안 지역으로는 최대 30cm 이상, 제주 산지에는 최고 50cm 이상의 폭설이 우려됩니다. 충남 지역에는 금요일까지 호남과 제주도는 일요일까지 눈이 길게 이어지겠습니다. 날씨였습니다. https://imnews.imbc.com/replay/2021/nw1700/article/6048848_34922.html #날씨,#추위 #한파",
"video_id": "QmU3W9ffZeM"
}
| 4
|
[
"The video shows a weather forecast for Seoul, highlighting a drop in temperatures, with a woman in a maroon dress presenting the information. The graph indicates varying temperatures over the day, starting at -5.9°C and reaching as low as -17°C. The background shows a river with some grass, indicating cold weather.",
"The video features a woman in a maroon dress presenting weather information, pointing at a map indicating regions affected by the weather. The map highlights the Han River Basin in blue, signifying cold temperatures. The background shows a river with grass, reinforcing the cold weather theme.",
"The video continues with the woman in a maroon dress presenting weather information. She stands outdoors against a snowy backdrop with trees and a river visible. Text on the screen indicates the weather forecast predicts snowfall of up to 30 cm starting tonight. The woman appears to be explaining the forecast, emphasizing the expected heavy snowfall.",
"The woman in a maroon dress continues her weather presentation outdoors. She is explaining that heavy snowfall of up to 30 cm is expected starting tonight. The backdrop remains snowy with trees and a river visible. She uses hand gestures to emphasize the forecast, indicating the significance of the upcoming weather conditions."
] |
QmU3W9ffZeM
|
A weather forecast video for Seoul features a woman in a maroon dress presenting information about an impending cold snap. The forecast indicates a significant drop in temperatures, ranging from -5.9°C to as low as -17°C, with the Han River Basin expected to be particularly affected. The presenter emphasizes the prediction of heavy snowfall, up to 30 cm, starting tonight, using hand gestures and standing against a snowy outdoor backdrop with trees and a river visible, to convey the importance of the upcoming weather conditions.
|
In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.
|
[
"People bundling up in heavy clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in heavy clothing to brave the cold during record cold temperatures in East Asia, Southeast Asia, and South Asia in late January 2016",
"People bundling up in heavy clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong, during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia during late January 2016 snowfall events",
"Frost or ice formation on car windshields, roads, and buildings during the cold wave in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings during snowfall in East Asia, Southeast Asia, and South Asia in late January 2016",
"Residents in East Asia, Southeast Asia, and South Asia shoveling snow or using umbrellas to navigate snowy streets during the late January 2016 cold wave",
"Residents in East and South Asia shoveling snow or using umbrellas to navigate snowy streets during record cold temperatures in late January 2016",
"Residents shoveling snow or using umbrellas to navigate snowy streets in East Asia, Southeast Asia, or South Asia in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia Southeast Asia and South Asia during late January 2016"
] |
[
"People bundling up in warm clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in warm clothing to brave the cold in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People bundling up in warm clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"Frozen lakes and rivers with people ice skating or playing in East Asia Southeast Asia or South Asia during late January 2016 cold wave",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia where people are ice skating or playing during the record cold temperatures of late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, or South Asia where people are ice skating or playing in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia during late January 2016 cold wave",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia after snowfall in late January 2016",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 cold wave",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off sidewalks and driveways in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
[
"People bundling up in winter clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in winter clothing to brave the cold in East and South Asia during late January 2016 record cold snap",
"People bundling up in winter clothing to brave the cold during late January 2016 snowfall in East, Southeast, and South Asia",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during the cold wave of late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia due to record cold temperatures in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia after snowfall in late January 2016",
"Trees and buildings covered in snow and ice during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off streets and sidewalks in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_1%EC%9B%94_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%95%9C%ED%8C%8C",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, and the consequences of the snowfall, as these events are likely to be visualized in a video. The extracted events should be specific to the regions mentioned and the time of year, taking into account the typical climate and scenery of East Asia, mainland Southeast Asia, and northern South Asia in late January.\n\nEVENTS:\n1. People bundling up in winter clothing to brave the cold\n2. Snow falling in cities and towns\n3. Frozen lakes and rivers\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off streets and sidewalks",
"en_article_excerpt": "In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.",
"en_article_url": "https://en.wikipedia.org/wiki/January_2016_East_Asia_cold_wave",
"event_category": "disasters",
"event_name": "2016_east_asia_cold_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, disruptions to daily life, and unusual weather phenomena like snowfall in regions that rarely see it. These events would be visually representative of the cold wave's impact and could be reasonably captured in a video.\n\nEVENTS:\n1. People bundling up in heavy clothing to brave the cold\n2. Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong\n3. Frost or ice forming on surfaces like car windshields, roads, and buildings\n4. Residents shoveling snow or using umbrellas to navigate snowy streets\n5. Cars and buses driving slowly on icy or snow-covered roads",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a cold wave, such as people dealing with the cold and snow, and disruptions to daily life. These events should be visual and concrete, and could be reasonably captured in a video of the cold wave's impact.\n\nEVENTS:\n1. People bundling up in warm clothing to brave the cold\n2. Snowplows clearing roads and highways\n3. Frozen lakes and rivers with people ice skating or playing\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off sidewalks and driveways",
"video_URL": "http://youtube.com/watch?v=G7XGz4uL6mI",
"video_description": "기상정보 배혜지 기상캐스터 2021년 1월 12일 화요일 아침 시각 빙판길 골목길 언덕길 등 보행 조심하세요... 감기 걸리지 않도록 건강관리 잘하세요. 행복한 화요일 되세요.^^ #감기조심 #한파특보 #한파경보 #한파주의보 #건조경보 #화재예방 #산불조심 #코로나19 #COVID19 #기상정보 #날씨 #Weather #배혜지기상캐스터 #기상캐스터 #WeatherCaster #KBS뉴스광장 ============================================================== Uploaded by: 하안마음 (White Heart.rhythm7777777) Youtube: http://www.youtube.com/user/rhythmsgarden Twitter: https://twitter.com/rhythm700 Face Book: https://www.facebook.com/rhythm700 Instagram: https://www.instagram.com/rhythm7777777 헤일리 웨스튼라 Hayley Westenra Official Website: http://www.hayleywestenra.com Instagram: https://www.instagram.com/hayleywestenra Twitter: https://twitter.com/hayleywestenra FaceBook: https://www.facebook.com/hayleywestenra Youtube: https://www.youtube.com/channel/UCuA623UAgxrJuWcMcaI_kgg 헤일리 웨스튼라 코리아 Hayley Westenra Korea Fancafe: http://cafe.daum.net/HayleyWestenra Facebook: https://www.facebook.com/hayleywestenra Youtube: https://www.youtube.com/channel/UCBrwEte0dnl1JTWSpzGFVeA/videos Wikipedia: http://en.wikipedia.org/wiki/Hayley_Westenra",
"video_id": "G7XGz4uL6mI"
}
| 4
|
[
"A weather presenter stands in front of a snowy mountain backdrop, delivering a forecast. A graphic on the screen displays temperature comparisons for the morning and today versus the previous year for two locations: 아침 (Morning) and 낮 (Day). The graphics show that temperatures are slightly higher this year compared to the previous year. The text indicates the current time as 12th at 7:47.",
"A weather presenter stands in front of a map of South Korea, gesturing towards it. The map highlights specific regions, with labels indicating areas of interest. The presenter, wearing a yellow dress, continues to deliver the forecast, likely explaining the weather conditions or forecasts for these regions. The time displayed remains 12th at 7:47.",
"The weather presenter continues to stand in front of the map, still pointing out areas of interest. The map now includes additional times labeled \"12시~15시,\" \"15시~18시,\" and \"18시~21시,\" suggesting detailed weather updates for different time slots. The presenter, still in the yellow dress, continues the forecast, providing updates for the highlighted regions. The time displayed has changed to 12th at 7:48, indicating the progression of the video.",
"The weather presenter stands in front of a snowy background, continuing the forecast. The map on the screen shows areas of interest with times labeled \"12시~15시,\" \"15시~18시,\" and \"18시~21시,\" indicating detailed weather updates for different time slots. The presenter, still in the yellow dress, appears to be providing updates for the highlighted regions. The time displayed is 12th at 7:48, consistent with the previous frame."
] |
G7XGz4uL6mI
|
A weather presenter, wearing a yellow dress, delivers a forecast in front of a snowy mountain backdrop and a map of South Korea. The presenter compares temperatures from the current year to the previous year, showing a slight increase, and provides detailed weather updates for specific regions of South Korea at different time slots (12시~15시, 15시~18시, and 18시~21시). The forecast is presented over a period of at least one minute, from 7:47 to 7:48 on the 12th.
|
In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.
|
[
"People bundling up in heavy clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in heavy clothing to brave the cold during record cold temperatures in East Asia, Southeast Asia, and South Asia in late January 2016",
"People bundling up in heavy clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong, during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia during late January 2016 snowfall events",
"Frost or ice formation on car windshields, roads, and buildings during the cold wave in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings during snowfall in East Asia, Southeast Asia, and South Asia in late January 2016",
"Residents in East Asia, Southeast Asia, and South Asia shoveling snow or using umbrellas to navigate snowy streets during the late January 2016 cold wave",
"Residents in East and South Asia shoveling snow or using umbrellas to navigate snowy streets during record cold temperatures in late January 2016",
"Residents shoveling snow or using umbrellas to navigate snowy streets in East Asia, Southeast Asia, or South Asia in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia Southeast Asia and South Asia during late January 2016"
] |
[
"People bundling up in warm clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in warm clothing to brave the cold in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People bundling up in warm clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"Frozen lakes and rivers with people ice skating or playing in East Asia Southeast Asia or South Asia during late January 2016 cold wave",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia where people are ice skating or playing during the record cold temperatures of late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, or South Asia where people are ice skating or playing in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia during late January 2016 cold wave",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia after snowfall in late January 2016",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 cold wave",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off sidewalks and driveways in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
[
"People bundling up in winter clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in winter clothing to brave the cold in East and South Asia during late January 2016 record cold snap",
"People bundling up in winter clothing to brave the cold during late January 2016 snowfall in East, Southeast, and South Asia",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during the cold wave of late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia due to record cold temperatures in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia after snowfall in late January 2016",
"Trees and buildings covered in snow and ice during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off streets and sidewalks in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_1%EC%9B%94_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%95%9C%ED%8C%8C",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, and the consequences of the snowfall, as these events are likely to be visualized in a video. The extracted events should be specific to the regions mentioned and the time of year, taking into account the typical climate and scenery of East Asia, mainland Southeast Asia, and northern South Asia in late January.\n\nEVENTS:\n1. People bundling up in winter clothing to brave the cold\n2. Snow falling in cities and towns\n3. Frozen lakes and rivers\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off streets and sidewalks",
"en_article_excerpt": "In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.",
"en_article_url": "https://en.wikipedia.org/wiki/January_2016_East_Asia_cold_wave",
"event_category": "disasters",
"event_name": "2016_east_asia_cold_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, disruptions to daily life, and unusual weather phenomena like snowfall in regions that rarely see it. These events would be visually representative of the cold wave's impact and could be reasonably captured in a video.\n\nEVENTS:\n1. People bundling up in heavy clothing to brave the cold\n2. Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong\n3. Frost or ice forming on surfaces like car windshields, roads, and buildings\n4. Residents shoveling snow or using umbrellas to navigate snowy streets\n5. Cars and buses driving slowly on icy or snow-covered roads",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a cold wave, such as people dealing with the cold and snow, and disruptions to daily life. These events should be visual and concrete, and could be reasonably captured in a video of the cold wave's impact.\n\nEVENTS:\n1. People bundling up in warm clothing to brave the cold\n2. Snowplows clearing roads and highways\n3. Frozen lakes and rivers with people ice skating or playing\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off sidewalks and driveways",
"video_URL": "http://twitter.com/i/status/1346762056142188546",
"video_description": "눈보라 때문에 기절을 함",
"video_id": "1346762056142188546"
}
| 4
|
[
"A snowstorm is occurring in an urban area, with snow heavily falling and covering the streets. The scene shows storefronts with neon signs, including one that reads \"JUNO HAIR\" and another with Korean text.",
"The snowstorm continues in the urban area, with heavy snowfall persistently covering the streets. Storefronts with neon signs, including one reading \"JUNO HAIR\" and another with Korean text, are still visible, and the scene remains consistently covered in snow.",
"The heavy snowfall persists in the urban area, continuing to blanket the streets. Storefronts with neon signs, including one reading \"JUNO HAIR\" and another with Korean text, remain visible, and the scene is consistently covered in snow.",
"The snowfall continues to heavily cover the urban streets, maintaining the dense blanket of snow. The \"JUNO HAIR\" sign and the Korean text sign are still visible, indicating the ongoing visibility of storefronts despite the snow. A figure dressed in dark clothing is walking away from the camera, navigating the snow-covered street, emphasizing the persistence of daily activities in the snowy conditions."
] |
1346762056142188546
|
A snowstorm is occurring in an urban area with storefronts, including a hair salon called "JUNO HAIR" and another with Korean text, as the backdrop. Despite the heavy and persistent snowfall, a figure in dark clothing is seen walking away from the camera, navigating the snow-covered streets, indicating that daily activities continue amidst the storm. The scene remains consistently covered in snow throughout, with the storefronts and neon signs remaining visible throughout the duration.
|
In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.
|
[
"People bundling up in heavy clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in heavy clothing to brave the cold during record cold temperatures in East Asia, Southeast Asia, and South Asia in late January 2016",
"People bundling up in heavy clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong, during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia during late January 2016 snowfall events",
"Frost or ice formation on car windshields, roads, and buildings during the cold wave in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings during snowfall in East Asia, Southeast Asia, and South Asia in late January 2016",
"Residents in East Asia, Southeast Asia, and South Asia shoveling snow or using umbrellas to navigate snowy streets during the late January 2016 cold wave",
"Residents in East and South Asia shoveling snow or using umbrellas to navigate snowy streets during record cold temperatures in late January 2016",
"Residents shoveling snow or using umbrellas to navigate snowy streets in East Asia, Southeast Asia, or South Asia in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia Southeast Asia and South Asia during late January 2016"
] |
[
"People bundling up in warm clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in warm clothing to brave the cold in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People bundling up in warm clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"Frozen lakes and rivers with people ice skating or playing in East Asia Southeast Asia or South Asia during late January 2016 cold wave",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia where people are ice skating or playing during the record cold temperatures of late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, or South Asia where people are ice skating or playing in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia during late January 2016 cold wave",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia after snowfall in late January 2016",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 cold wave",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off sidewalks and driveways in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
[
"People bundling up in winter clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in winter clothing to brave the cold in East and South Asia during late January 2016 record cold snap",
"People bundling up in winter clothing to brave the cold during late January 2016 snowfall in East, Southeast, and South Asia",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during the cold wave of late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia due to record cold temperatures in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia after snowfall in late January 2016",
"Trees and buildings covered in snow and ice during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off streets and sidewalks in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_1%EC%9B%94_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%95%9C%ED%8C%8C",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, and the consequences of the snowfall, as these events are likely to be visualized in a video. The extracted events should be specific to the regions mentioned and the time of year, taking into account the typical climate and scenery of East Asia, mainland Southeast Asia, and northern South Asia in late January.\n\nEVENTS:\n1. People bundling up in winter clothing to brave the cold\n2. Snow falling in cities and towns\n3. Frozen lakes and rivers\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off streets and sidewalks",
"en_article_excerpt": "In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.",
"en_article_url": "https://en.wikipedia.org/wiki/January_2016_East_Asia_cold_wave",
"event_category": "disasters",
"event_name": "2016_east_asia_cold_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, disruptions to daily life, and unusual weather phenomena like snowfall in regions that rarely see it. These events would be visually representative of the cold wave's impact and could be reasonably captured in a video.\n\nEVENTS:\n1. People bundling up in heavy clothing to brave the cold\n2. Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong\n3. Frost or ice forming on surfaces like car windshields, roads, and buildings\n4. Residents shoveling snow or using umbrellas to navigate snowy streets\n5. Cars and buses driving slowly on icy or snow-covered roads",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a cold wave, such as people dealing with the cold and snow, and disruptions to daily life. These events should be visual and concrete, and could be reasonably captured in a video of the cold wave's impact.\n\nEVENTS:\n1. People bundling up in warm clothing to brave the cold\n2. Snowplows clearing roads and highways\n3. Frozen lakes and rivers with people ice skating or playing\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off sidewalks and driveways",
"video_URL": "http://twitter.com/i/status/1346961474866925570",
"video_description": "눈보라가 휘날리는 바람 찬 부산 시청 #풍산",
"video_id": "1346961474866925570"
}
| 4
|
[
"A person is shoveling snow on a sidewalk next to a parked van. Nearby, there are two other individuals observing the snow removal. The setting appears to be a city street with trees and a pedestrian crossing in the background.",
"In the current frame, several individuals are standing on a city sidewalk, some holding signs with text written in a foreign language. The scene continues to show a cold, wintry setting with snow on the ground and bare trees lining the street. A pedestrian crossing is visible in the foreground, and high-rise buildings can be seen in the distance. The individuals appear to be participating in an organized event, possibly a demonstration or public gathering.",
"In this frame, the focus remains on the city sidewalk where a person is standing, holding a smartphone to their ear. The wintry setting persists, with snow visible on the ground and leafless trees along the street. The pedestrian crossing is still prominent in the foreground, and the high-rise buildings in the distance provide a backdrop to the urban scene. This individual appears to be engaged in a phone conversation, possibly related to the ongoing event or gathering described earlier.",
"The video continues to depict a city street scene. The focus is on a white van with its rear doors open, parked on the side of the road, likely indicating that the occupants are unloading or loading items. The pedestrian crossing is still visible in the foreground, with several orange traffic cones placed around the area, suggesting some form of roadwork or traffic management. The leafless trees and wintry setting are consistent with the previous frames, maintaining the urban wintertime atmosphere."
] |
1346961474866925570
|
A group of individuals gather on a city sidewalk in a wintry setting, participating in an organized event, possibly a demonstration or public gathering. The scene unfolds with one person shoveling snow, while others observe or hold signs. As the event progresses, a person is seen engaged in a phone conversation, suggesting coordination or communication related to the gathering. Meanwhile, a white van is parked nearby, with its rear doors open, indicating the loading or unloading of items, and traffic cones are placed around the area to manage traffic. The overall context suggests a planned, possibly community-oriented event taking place in an urban environment during winter.
|
In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.
|
[
"People bundling up in heavy clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in heavy clothing to brave the cold during record cold temperatures in East Asia, Southeast Asia, and South Asia in late January 2016",
"People bundling up in heavy clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong, during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snowflakes falling in cities like Bangkok or Hong Kong in East Asia during late January 2016 snowfall events",
"Frost or ice formation on car windshields, roads, and buildings during the cold wave in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Frost or ice forming on surfaces like car windshields, roads, and buildings during snowfall in East Asia, Southeast Asia, and South Asia in late January 2016",
"Residents in East Asia, Southeast Asia, and South Asia shoveling snow or using umbrellas to navigate snowy streets during the late January 2016 cold wave",
"Residents in East and South Asia shoveling snow or using umbrellas to navigate snowy streets during record cold temperatures in late January 2016",
"Residents shoveling snow or using umbrellas to navigate snowy streets in East Asia, Southeast Asia, or South Asia in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Cars and buses driving slowly on icy or snow-covered roads in East Asia Southeast Asia and South Asia during late January 2016"
] |
[
"People bundling up in warm clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in warm clothing to brave the cold in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People bundling up in warm clothing to brave the cold during late January 2016 snowfall in East Asia, Southeast Asia, and South Asia",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"Snowplows clearing roads and highways in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"Frozen lakes and rivers with people ice skating or playing in East Asia Southeast Asia or South Asia during late January 2016 cold wave",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia where people are ice skating or playing during the record cold temperatures of late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, or South Asia where people are ice skating or playing in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia during late January 2016 cold wave",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia Southeast Asia and South Asia after snowfall in late January 2016",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 cold wave",
"People shoveling snow off sidewalks and driveways in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off sidewalks and driveways in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
[
"People bundling up in winter clothing to brave the cold during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"People bundling up in winter clothing to brave the cold in East and South Asia during late January 2016 record cold snap",
"People bundling up in winter clothing to brave the cold during late January 2016 snowfall in East, Southeast, and South Asia",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during the cold wave of late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Snow falling in cities and towns in East Asia, Southeast Asia, and South Asia in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia due to record cold temperatures in late January 2016",
"Frozen lakes and rivers in East Asia, Southeast Asia, and South Asia after snowfall in late January 2016",
"Trees and buildings covered in snow and ice during the late January 2016 cold wave in East Asia, Southeast Asia, and South Asia",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during record cold temperatures in late January 2016",
"Trees and buildings covered in snow and ice in East Asia, Southeast Asia, and South Asia during late January 2016 snowfall",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during the cold wave in late January 2016",
"People shoveling snow off streets and sidewalks in East Asia, Southeast Asia, and South Asia during late January 2016 record cold temperatures",
"People shoveling snow off streets and sidewalks in East Asia Southeast Asia and South Asia during late January 2016 snowfall"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_1%EC%9B%94_%EB%8F%99%EC%95%84%EC%8B%9C%EC%95%84_%ED%95%9C%ED%8C%8C",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, and the consequences of the snowfall, as these events are likely to be visualized in a video. The extracted events should be specific to the regions mentioned and the time of year, taking into account the typical climate and scenery of East Asia, mainland Southeast Asia, and northern South Asia in late January.\n\nEVENTS:\n1. People bundling up in winter clothing to brave the cold\n2. Snow falling in cities and towns\n3. Frozen lakes and rivers\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off streets and sidewalks",
"en_article_excerpt": "In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.",
"en_article_url": "https://en.wikipedia.org/wiki/January_2016_East_Asia_cold_wave",
"event_category": "disasters",
"event_name": "2016_east_asia_cold_wave",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of the cold wave, such as people experiencing the cold weather, disruptions to daily life, and unusual weather phenomena like snowfall in regions that rarely see it. These events would be visually representative of the cold wave's impact and could be reasonably captured in a video.\n\nEVENTS:\n1. People bundling up in heavy clothing to brave the cold\n2. Snowflakes falling in cities that rarely see snow, such as Bangkok or Hong Kong\n3. Frost or ice forming on surfaces like car windshields, roads, and buildings\n4. Residents shoveling snow or using umbrellas to navigate snowy streets\n5. Cars and buses driving slowly on icy or snow-covered roads",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: In late January 2016, a cold wave struck much of East Asia, parts of mainland Southeast Asia and parts of northern South Asia, bringing record cold temperatures and snowfall to many regions.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the effects of a cold wave, such as people dealing with the cold and snow, and disruptions to daily life. These events should be visual and concrete, and could be reasonably captured in a video of the cold wave's impact.\n\nEVENTS:\n1. People bundling up in warm clothing to brave the cold\n2. Snowplows clearing roads and highways\n3. Frozen lakes and rivers with people ice skating or playing\n4. Trees and buildings covered in snow and ice\n5. People shoveling snow off sidewalks and driveways",
"video_URL": "http://youtube.com/watch?v=PxIbNGPQYw8",
"video_description": "지난주 부터 계속되는 한파에 SNS 등에서는 순간포착 영상인 이른바 '얼었어' 시리즈가 유행하고 있다고 합니다. https://imnews.imbc.com/replay/2021/nwtoday/article/6054105_34943.html #한파, #SNS, #얼었어시리즈",
"video_id": "PxIbNGPQYw8"
}
| 4
|
[
"The image shows two men in suits standing on a news set, engaged in a discussion. The backdrop features a cityscape and text in Korean, indicating that the scene is part of a news broadcast on MBC, dated 11 (월) 7:16. The text at the top of the screen reads \"뉴스터치\" (News Touch), and below, there is a segment title \"한파 이겨내는 '열었어' 시리즈,\" which translates to \"Series on 'Opened' Against the Cold Wave.\" Additional text discusses the impact of cold weather and car sales, mentioning terms like \"충고차 시장\" (Advisory Car Market) and \"차식 인구 10년 새 10배 증가\" (Car population increased tenfold in ten years).",
"The current frame shows a close-up of a hand opening a package of sanitary wipes, a plastic bottle of water, and a bottle of Coca-Cola Zero Sugar. The scene transitions from the previous frame's news broadcast, continuing the discussion on consumer behavior and product usage during cold weather. The text \"최강 한파에 '열었어' 시리즈 등장\" translates to \"Series on 'Opened' Against the Strongest Cold Wave.\" The display of everyday products, such as wipes and beverages, highlights the practical items people are using or considering during the cold season, as mentioned in the segment title and the reference to a $160,000 bonus for the KPX group.",
"The current frame shows a close-up of a faucet with water flowing out, alongside a bowl of noodles being lifted with chopsticks. The text \"전국 계량기 동파 사고 5,000여 건\" translates to \"Over 5,000 Frost Damage Cases to Meters Nationwide.\" The scene relates to the news segment about the impact of the cold weather on household utilities, specifically highlighting the damage to water meters due to freezing temperatures. The display of the faucet and noodles might be illustrating the necessity of maintaining warm water flow and the importance of heating food during such weather conditions.",
"The current frame continues to highlight the impact of freezing weather on daily life, featuring a close-up of a faucet with water flowing out, alongside a bowl of noodles being lifted with chopsticks. The text \"전국 계량기 동파 사고 5,000여 건\" translates to \"Over 5,000 Frost Damage Cases to Meters Nationwide,\" emphasizing the extensive damage to water meters due to cold temperatures. The display of the faucet and noodles suggests the importance of maintaining water flow and the need for heating food, illustrating the challenges faced during such weather conditions."
] |
PxIbNGPQYw8
|
A news broadcast on MBC, titled "News Touch," discusses the impact of a cold wave on daily life, featuring a series called "Opened" that explores how people cope with the cold weather. The segment touches on the increased demand for certain products, such as sanitary wipes and beverages, and highlights the challenges faced by households, including damage to water meters due to freezing temperatures, with over 5,000 cases reported nationwide. The broadcast also mentions the growth of the car market and the increase in car population over the past decade. Overall, the video summarizes the effects of the cold wave on consumer behavior, household utilities, and daily life in general.
|
Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.
|
[
"Heavy rains and floods in the southern Korean Peninsula during summer 2020",
"Heavy rains pouring down in southern South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North and South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during the summer floods of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during heavy rains in summer 2020",
"Rescue teams navigating boats through flooded streets to reach stranded residents during the summer floods of 2020 in North and South Korea",
"Rescue teams navigating boats through flooded streets to reach stranded residents in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after floods in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after heavy rains in North and South Korea during the summer of 2020"
] |
[
"People evacuating flooded homes and buildings in the Korean Peninsula during the summer rainy season of 2020",
"People evacuating flooded homes and buildings in the Korean Peninsula during summer 2020 heavy rains",
"Cars and buildings submerged in floodwater in North Korea and South Korea during the summer floods of 2020",
"Cars and buildings submerged in floodwater during heavy rains in the Korean Peninsula in summer 2020",
"Emergency responders rescuing people from flooded areas in North Korea and South Korea during the summer rainy season of 2020",
"Emergency responders rescuing people from flooded areas in North and South Korea during heavy rains in the summer of 2020",
"Debris and wreckage from destroyed buildings and infrastructure due to floods in North and South Korea during the summer 2020 rainy season",
"Debris and wreckage from destroyed buildings and infrastructure in North Korea, South Korea, and the Korean Peninsula after heavy rains during the summer rainy season of June-August 2020",
"Residents clearing mud and water from their homes and streets after floods in the Korean Peninsula during the summer 2020 rainy season",
"Residents clearing mud and water from their homes and streets in North and South Korea after heavy rains during the summer rainy season of 2020"
] |
[
"Floodwaters inundating homes and buildings in southern Korea, particularly in South Korea, during the summer rainy season of 2020",
"Floodwaters inundating homes and buildings in southern Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North and South Korea during heavy rains in summer 2020",
"Rescue workers helping those stranded by the floods in North Korea and South Korea during the summer rainy season of 2020",
"Rescue workers helping those stranded by floods in the Korean Peninsula during the summer rainy season of 2020",
"Cars and buildings submerged in water during floods in North and South Korea, Korean Peninsula, Summer 2020",
"Cars and buildings submerged in water in North and South Korea during heavy rains in summer 2020",
"Emergency responders distributing aid and supplies to communities affected by floods in North Korea and South Korea during the summer rainy season of June to August 2020",
"Emergency responders distributing aid and supplies to affected communities in North and South Korea during heavy rains in the summer of 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2020%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact of floods and heavy rains in Korea, such as scenes of flooding, damage, and rescue efforts, as these are likely to be visualized in a video about the event. The extracted events should be specific to the region and the time of year, taking into account the regional rainy season and the areas most affected by the floods.\n\nEVENTS:\n1. Floodwaters inundating homes and buildings in southern Korea\n2. People evacuating flooded areas by boat\n3. Rescue workers helping those stranded by the floods\n4. Cars and buildings submerged in water\n5. Emergency responders distributing aid and supplies to affected communities",
"en_article_excerpt": "Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Korean_floods",
"event_category": "disasters",
"event_name": "2020_korean_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the cause and effects of the flooding, such as scenes of heavy rainfall, flooded areas, and rescue efforts. These events are likely to be visualized in a video documenting the floods and their impact on the region.\n\nEVENTS:\n1. Heavy rains pouring down in the southern parts of the Korean Peninsula\n2. Rivers overflowing and flooding nearby cities and towns\n3. People evacuating their homes and seeking higher ground\n4. Rescue teams navigating boats through flooded streets to reach stranded residents\n5. Cars and buildings submerged in water with debris scattered around",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the consequences of flooding and heavy rainfall, such as damage to infrastructure, evacuation efforts, and scenes of flooding. These events are likely to be visualized in a video capturing the impact of the floods on the region.\n\nEVENTS:\n1. People evacuating flooded homes and buildings\n2. Cars and buildings submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Debris and wreckage from destroyed buildings and infrastructure\n5. Residents clearing mud and water from their homes and streets",
"video_URL": "http://youtube.com/watch?v=SPZkwqJerMU",
"video_description": "2020년 홍수 피해 주민들에 1천483억원 보상 2020년 여름 기나긴 장마로 큰 홍수 피해를 당한 주민들이 총 1천480여억원의 보상금을 지급받게 됐습니다. 환경부 소속 중앙환경분쟁조정위원회는 홍수 피해를 본 주민들과 정부, 지자체 및 한국수자원공사 간 분쟁조정 절차가 최근 마무리됐다고 밝혔습니다. 조정위는 관계부처 원인 조사에서 댐·하천을 미흡하게 관리한 사실이 드러난 점 등을 고려했다고 밝혔습니다. 연합뉴스TV 기사문의 및 제보 : 카톡/라인 jebo23 ▣ 연합뉴스TV 유튜브 채널 구독 https://goo.gl/VuCJMi ▣ 대한민국 뉴스의 시작 연합뉴스TV / Yonhap News TV http://www.yonhapnewstv.co.kr/",
"video_id": "SPZkwqJerMU"
}
| 4
|
[
"The image shows a large pile of assorted trash, including cardboard, plastic bags, and other waste materials, on a muddy surface. There is a large number of yellow bags among the debris. In the background, a river is visible. At the top right, there is Korean text indicating \"연합뉴스 TV\" and \"신규확진\" with the number \"서울 65,033,\" suggesting a news report about a large trash pile and a significant number of new cases in Seoul. The bottom banner mentions \"2020년 홍수 피해 주민들에 1천483억원 보상,\" indicating a compensation of 1.483 billion won for flood victims in 2020.",
"The current frame shows a close-up view of the same large pile of assorted trash, including cardboard, plastic bags, and other waste materials, on a muddy surface. The yellow bags from the previous frame are still visible among the debris. There is no significant change in the scene, maintaining the focus on the trash pile. In the background, a river is still visible. The Korean text at the top right continues to read \"연합뉴스 TV\" and \"신규확진\" with the number \"서울 65,033,\" reinforcing the news report about the large trash pile and the significant number of new cases in Seoul. The bottom banner still mentions \"2020년 홍수 피해 주민들에 1천483억원 보상,\" indicating the ongoing coverage of the 1.483 billion won compensation for flood victims in 2020.",
"The current frame captures a scene of significant destruction, likely due to a natural disaster, with a large pile of debris prominently in the foreground. The debris includes various materials such as wood, metal, and other unidentifiable items, indicating extensive damage. In the background, there is a building with a partially collapsed roof, suggesting the force of the event that led to this destruction. The Korean text at the top of the frame reads \"연합뉴스 TV\" and \"신규확진\" with the number \"인천 17,568,\" which indicates a news report from Incheon with a certain number of new cases. The bottom banner still mentions \"2020년 홍수 피해 주민들에 1천483억원 보상,\" indicating ongoing coverage of the compensation for flood victims in 2020. The scene continues to highlight the aftermath of a disaster, focusing on the immediate effects on the environment and infrastructure.",
"The current frame shows a scene that appears to be in a state of disarray, likely in the aftermath of a disaster. The foreground features a large pile of debris, which includes various materials such as wood, metal, and other unidentifiable items, indicating extensive damage. In the background, there is a building with a partially collapsed roof, suggesting the force of the event that led to this destruction. The Korean text at the top of the frame reads \"연합뉴스 TV\" and \"신규확진\" with the number \"인천 17,568,\" which indicates a news report from Incheon with a certain number of new cases. The bottom banner mentions \"2020년 홍수 피해 주민들에 1천483억원 보상,\" indicating ongoing coverage of the compensation for flood victims in 2020. The scene continues to highlight the aftermath of a disaster, focusing on the immediate effects on the environment and infrastructure."
] |
SPZkwqJerMU
|
The video appears to be a news report about the aftermath of a significant flood in South Korea, specifically highlighting the damage and debris left behind. The scenes show large piles of trash and debris, including cardboard, plastic bags, wood, and metal, scattered across muddy surfaces with a river visible in the background. The report also mentions a large number of new cases, potentially referring to COVID-19 cases, with specific numbers given for Seoul and Incheon. Additionally, the video covers the compensation of 1.483 billion won for flood victims in 2020, indicating that the flood occurred that year. The overall context suggests a news broadcast covering both the environmental disaster and its impact on the community, as well as the ongoing pandemic.
|
Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.
|
[
"Heavy rains and floods in the southern Korean Peninsula during summer 2020",
"Heavy rains pouring down in southern South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North and South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during the summer floods of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during heavy rains in summer 2020",
"Rescue teams navigating boats through flooded streets to reach stranded residents during the summer floods of 2020 in North and South Korea",
"Rescue teams navigating boats through flooded streets to reach stranded residents in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after floods in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after heavy rains in North and South Korea during the summer of 2020"
] |
[
"People evacuating flooded homes and buildings in the Korean Peninsula during the summer rainy season of 2020",
"People evacuating flooded homes and buildings in the Korean Peninsula during summer 2020 heavy rains",
"Cars and buildings submerged in floodwater in North Korea and South Korea during the summer floods of 2020",
"Cars and buildings submerged in floodwater during heavy rains in the Korean Peninsula in summer 2020",
"Emergency responders rescuing people from flooded areas in North Korea and South Korea during the summer rainy season of 2020",
"Emergency responders rescuing people from flooded areas in North and South Korea during heavy rains in the summer of 2020",
"Debris and wreckage from destroyed buildings and infrastructure due to floods in North and South Korea during the summer 2020 rainy season",
"Debris and wreckage from destroyed buildings and infrastructure in North Korea, South Korea, and the Korean Peninsula after heavy rains during the summer rainy season of June-August 2020",
"Residents clearing mud and water from their homes and streets after floods in the Korean Peninsula during the summer 2020 rainy season",
"Residents clearing mud and water from their homes and streets in North and South Korea after heavy rains during the summer rainy season of 2020"
] |
[
"Floodwaters inundating homes and buildings in southern Korea, particularly in South Korea, during the summer rainy season of 2020",
"Floodwaters inundating homes and buildings in southern Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North and South Korea during heavy rains in summer 2020",
"Rescue workers helping those stranded by the floods in North Korea and South Korea during the summer rainy season of 2020",
"Rescue workers helping those stranded by floods in the Korean Peninsula during the summer rainy season of 2020",
"Cars and buildings submerged in water during floods in North and South Korea, Korean Peninsula, Summer 2020",
"Cars and buildings submerged in water in North and South Korea during heavy rains in summer 2020",
"Emergency responders distributing aid and supplies to communities affected by floods in North Korea and South Korea during the summer rainy season of June to August 2020",
"Emergency responders distributing aid and supplies to affected communities in North and South Korea during heavy rains in the summer of 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2020%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact of floods and heavy rains in Korea, such as scenes of flooding, damage, and rescue efforts, as these are likely to be visualized in a video about the event. The extracted events should be specific to the region and the time of year, taking into account the regional rainy season and the areas most affected by the floods.\n\nEVENTS:\n1. Floodwaters inundating homes and buildings in southern Korea\n2. People evacuating flooded areas by boat\n3. Rescue workers helping those stranded by the floods\n4. Cars and buildings submerged in water\n5. Emergency responders distributing aid and supplies to affected communities",
"en_article_excerpt": "Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Korean_floods",
"event_category": "disasters",
"event_name": "2020_korean_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the cause and effects of the flooding, such as scenes of heavy rainfall, flooded areas, and rescue efforts. These events are likely to be visualized in a video documenting the floods and their impact on the region.\n\nEVENTS:\n1. Heavy rains pouring down in the southern parts of the Korean Peninsula\n2. Rivers overflowing and flooding nearby cities and towns\n3. People evacuating their homes and seeking higher ground\n4. Rescue teams navigating boats through flooded streets to reach stranded residents\n5. Cars and buildings submerged in water with debris scattered around",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the consequences of flooding and heavy rainfall, such as damage to infrastructure, evacuation efforts, and scenes of flooding. These events are likely to be visualized in a video capturing the impact of the floods on the region.\n\nEVENTS:\n1. People evacuating flooded homes and buildings\n2. Cars and buildings submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Debris and wreckage from destroyed buildings and infrastructure\n5. Residents clearing mud and water from their homes and streets",
"video_URL": "http://youtube.com/watch?v=mtt7-wXFwxI",
"video_description": "(서울=뉴스1) 김연수 기자 = 지난 1일부터 전국을 강타하고 있는 집중호우로 전국에서 피해가 속출하고 있습니다. 특히 큰 비가 내린 남부 지방을 중심으로 지난 7일 이후 1만1717건의 시설피해가 집계됐다. 섬진강댐 하류 제방이 붕괴돼 주민들이 긴급대피했고 화동군 화개장터가 32년 만에 침수됐습니다. 도로와 교량, 주택 등 1만7879건의 시설피해가 접수된 가운데 제5호 태풍 '장미'가 북상하면서 피해 규모가 더욱 커질 것으로 전망되는데요. 수마가 할퀴고 간 자리 영상으로 확인해보세요. #장마 #홍수피해 #홍수 #물폭탄",
"video_id": "mtt7-wXFwxI"
}
| 4
|
[
"The image shows a group of rescue workers navigating through a flooded street in a city. They are in a small inflatable boat, wearing red life jackets and helmets, indicating a rescue operation. The water level is high, reaching up to the knees of individuals walking through it. Buildings line the street, some with signs written in Korean, suggesting the location is in Korea. The rescue workers appear to be moving towards a person standing in the water, who is also wearing a helmet, likely coordinating the rescue efforts. The text on the screen includes \"영상출처 | 전남소방본부\" and \"뉴스1TV,\" indicating the source of the footage.",
"The current frame shows a flooded river with muddy water, and a single tree partially submerged in the water. The riverbank is visible on the right side of the frame, with a wet, paved path leading away from the water. A group of people is walking along the path, some holding umbrellas. The background features a city skyline with several high-rise buildings. The text on the screen includes \"영상출처 | 노정철 독자 제공\" and \"뉴스1TV,\" indicating the source of the footage.",
"The current frame shows a flooded river with muddy water, and a single tree partially submerged in the water. The riverbank is visible on the right side of the frame, with a wet, paved path leading away from the water. A group of people is walking along the path, some holding umbrellas. The background features a city skyline with several high-rise buildings. The text on the screen includes \"영상출처 | 인스타그램 @aitom__\" and \"뉴스1TV,\" indicating the source of the footage.",
"The current frame shows a blue gradient background with the text \"뉴스1TV\" prominently displayed in the center. Below the logo, there are additional texts in Korean: \"영상 | 전남소방본부 등\" on the left and \"제작 | 김연수\" on the right, suggesting that the footage is sourced from the Jeonnam Fire Department and produced by Kim Yeon-soo. The frame indicates a transition or a segment title within the video."
] |
mtt7-wXFwxI
|
A rescue operation is underway in a flooded city in Korea, with rescue workers navigating through the flooded streets in an inflatable boat to reach individuals in need. The flooding is severe, with water levels reaching up to the knees of those walking through it. The rescue efforts are coordinated, with individuals directing the operation from the water. The flooding also affects a nearby river, which is shown with muddy water and a partially submerged tree, and a group of people are seen walking along the riverbank, seeking higher ground. The video is a news report, sourced from various contributors, including the Jeonnam Fire Department, and produced by Kim Yeon-soo for 뉴스1TV.
|
Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.
|
[
"Heavy rains and floods in the southern Korean Peninsula during summer 2020",
"Heavy rains pouring down in southern South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North and South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during the summer floods of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during heavy rains in summer 2020",
"Rescue teams navigating boats through flooded streets to reach stranded residents during the summer floods of 2020 in North and South Korea",
"Rescue teams navigating boats through flooded streets to reach stranded residents in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after floods in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after heavy rains in North and South Korea during the summer of 2020"
] |
[
"People evacuating flooded homes and buildings in the Korean Peninsula during the summer rainy season of 2020",
"People evacuating flooded homes and buildings in the Korean Peninsula during summer 2020 heavy rains",
"Cars and buildings submerged in floodwater in North Korea and South Korea during the summer floods of 2020",
"Cars and buildings submerged in floodwater during heavy rains in the Korean Peninsula in summer 2020",
"Emergency responders rescuing people from flooded areas in North Korea and South Korea during the summer rainy season of 2020",
"Emergency responders rescuing people from flooded areas in North and South Korea during heavy rains in the summer of 2020",
"Debris and wreckage from destroyed buildings and infrastructure due to floods in North and South Korea during the summer 2020 rainy season",
"Debris and wreckage from destroyed buildings and infrastructure in North Korea, South Korea, and the Korean Peninsula after heavy rains during the summer rainy season of June-August 2020",
"Residents clearing mud and water from their homes and streets after floods in the Korean Peninsula during the summer 2020 rainy season",
"Residents clearing mud and water from their homes and streets in North and South Korea after heavy rains during the summer rainy season of 2020"
] |
[
"Floodwaters inundating homes and buildings in southern Korea, particularly in South Korea, during the summer rainy season of 2020",
"Floodwaters inundating homes and buildings in southern Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North and South Korea during heavy rains in summer 2020",
"Rescue workers helping those stranded by the floods in North Korea and South Korea during the summer rainy season of 2020",
"Rescue workers helping those stranded by floods in the Korean Peninsula during the summer rainy season of 2020",
"Cars and buildings submerged in water during floods in North and South Korea, Korean Peninsula, Summer 2020",
"Cars and buildings submerged in water in North and South Korea during heavy rains in summer 2020",
"Emergency responders distributing aid and supplies to communities affected by floods in North Korea and South Korea during the summer rainy season of June to August 2020",
"Emergency responders distributing aid and supplies to affected communities in North and South Korea during heavy rains in the summer of 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2020%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact of floods and heavy rains in Korea, such as scenes of flooding, damage, and rescue efforts, as these are likely to be visualized in a video about the event. The extracted events should be specific to the region and the time of year, taking into account the regional rainy season and the areas most affected by the floods.\n\nEVENTS:\n1. Floodwaters inundating homes and buildings in southern Korea\n2. People evacuating flooded areas by boat\n3. Rescue workers helping those stranded by the floods\n4. Cars and buildings submerged in water\n5. Emergency responders distributing aid and supplies to affected communities",
"en_article_excerpt": "Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Korean_floods",
"event_category": "disasters",
"event_name": "2020_korean_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the cause and effects of the flooding, such as scenes of heavy rainfall, flooded areas, and rescue efforts. These events are likely to be visualized in a video documenting the floods and their impact on the region.\n\nEVENTS:\n1. Heavy rains pouring down in the southern parts of the Korean Peninsula\n2. Rivers overflowing and flooding nearby cities and towns\n3. People evacuating their homes and seeking higher ground\n4. Rescue teams navigating boats through flooded streets to reach stranded residents\n5. Cars and buildings submerged in water with debris scattered around",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the consequences of flooding and heavy rainfall, such as damage to infrastructure, evacuation efforts, and scenes of flooding. These events are likely to be visualized in a video capturing the impact of the floods on the region.\n\nEVENTS:\n1. People evacuating flooded homes and buildings\n2. Cars and buildings submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Debris and wreckage from destroyed buildings and infrastructure\n5. Residents clearing mud and water from their homes and streets",
"video_URL": "http://youtube.com/watch?v=5j_NI2eL3HA",
"video_description": "북한에서 집중 호우로 일어난 홍수에 22명이 숨지고 4명이 실종됐습니다. 국제적십자사연맹은 미국의소리 방송에 보낸 이메일에서 \"북한 조선 적십자회와 국가 비상 재해 위원회에 따르면 심각한 홍수로 인해 22명이 사망하고 4명이 실종됐다\"고 밝혔습니다. https://imnews.imbc.com/replay/2020/nwtoday/article/5873504_32531.html #북한 #집중호우 #홍수",
"video_id": "5j_NI2eL3HA"
}
| 4
|
[
"In the image, three people wearing dark raincoats are seen standing outdoors in a wooded area. One person is pointing or gesturing towards something off-frame. The background is lush with green trees, suggesting it might be raining. The text at the bottom of the image is in Korean, indicating some sort of news or documentary context.",
"A construction vehicle, specifically a bulldozer, is operating in a flooded area with high water levels. The bulldozer is equipped with a front loader, and it is being used to move debris and possibly build a makeshift barrier to control the floodwaters. The water appears muddy and turbulent, indicating severe flooding. Trees and vegetation are visible in the background, partially submerged in the floodwater. The text at the top of the image is in Korean, suggesting this is a news report. The image includes a timestamp and weather information, indicating it was captured at 7:24 AM on a day when the temperature is 26 degrees Celsius.",
"The video continues to capture the severe flooding situation. The bulldozer is still in action, moving through the flooded area with its front loader. The water remains muddy and turbulent, emphasizing the intensity of the flooding. Trees in the background are still partially submerged, indicating the extent of the flooding. The Korean text at the top of the image, along with the timestamp and weather information, suggests the footage is part of a news report, capturing the ongoing efforts to manage and mitigate the floodwaters at 7:24 AM with a temperature of 26 degrees Celsius.",
"The video continues to capture the severe flooding situation. The bulldozer is still in action, moving through the flooded area with its front loader. The water remains muddy and turbulent, emphasizing the intensity of the flooding. Trees in the background are still partially submerged, indicating the extent of the flooding. The Korean text at the top of the image, along with the timestamp and weather information, suggests the footage is part of a news report, capturing the ongoing efforts to manage and mitigate the floodwaters at 7:24 AM with a temperature of 26 degrees Celsius."
] |
5j_NI2eL3HA
|
A severe flooding event is depicted in the video, showing a wooded area overwhelmed by muddy and turbulent water. The footage, part of a news report, captures the efforts to manage and mitigate the floodwaters. A bulldozer equipped with a front loader is actively moving through the flooded area, likely to build a makeshift barrier or clear debris. The video begins with three people in raincoats standing in the wooded area, possibly discussing or assessing the situation. The main focus, however, is on the bulldozer's operations, highlighting the intensity of the flooding and the response efforts at 7:24 AM in 26-degree Celsius weather.
|
Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.
|
[
"Heavy rains and floods in the southern Korean Peninsula during summer 2020",
"Heavy rains pouring down in southern South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North and South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during the summer floods of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during heavy rains in summer 2020",
"Rescue teams navigating boats through flooded streets to reach stranded residents during the summer floods of 2020 in North and South Korea",
"Rescue teams navigating boats through flooded streets to reach stranded residents in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after floods in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after heavy rains in North and South Korea during the summer of 2020"
] |
[
"People evacuating flooded homes and buildings in the Korean Peninsula during the summer rainy season of 2020",
"People evacuating flooded homes and buildings in the Korean Peninsula during summer 2020 heavy rains",
"Cars and buildings submerged in floodwater in North Korea and South Korea during the summer floods of 2020",
"Cars and buildings submerged in floodwater during heavy rains in the Korean Peninsula in summer 2020",
"Emergency responders rescuing people from flooded areas in North Korea and South Korea during the summer rainy season of 2020",
"Emergency responders rescuing people from flooded areas in North and South Korea during heavy rains in the summer of 2020",
"Debris and wreckage from destroyed buildings and infrastructure due to floods in North and South Korea during the summer 2020 rainy season",
"Debris and wreckage from destroyed buildings and infrastructure in North Korea, South Korea, and the Korean Peninsula after heavy rains during the summer rainy season of June-August 2020",
"Residents clearing mud and water from their homes and streets after floods in the Korean Peninsula during the summer 2020 rainy season",
"Residents clearing mud and water from their homes and streets in North and South Korea after heavy rains during the summer rainy season of 2020"
] |
[
"Floodwaters inundating homes and buildings in southern Korea, particularly in South Korea, during the summer rainy season of 2020",
"Floodwaters inundating homes and buildings in southern Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North and South Korea during heavy rains in summer 2020",
"Rescue workers helping those stranded by the floods in North Korea and South Korea during the summer rainy season of 2020",
"Rescue workers helping those stranded by floods in the Korean Peninsula during the summer rainy season of 2020",
"Cars and buildings submerged in water during floods in North and South Korea, Korean Peninsula, Summer 2020",
"Cars and buildings submerged in water in North and South Korea during heavy rains in summer 2020",
"Emergency responders distributing aid and supplies to communities affected by floods in North Korea and South Korea during the summer rainy season of June to August 2020",
"Emergency responders distributing aid and supplies to affected communities in North and South Korea during heavy rains in the summer of 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2020%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact of floods and heavy rains in Korea, such as scenes of flooding, damage, and rescue efforts, as these are likely to be visualized in a video about the event. The extracted events should be specific to the region and the time of year, taking into account the regional rainy season and the areas most affected by the floods.\n\nEVENTS:\n1. Floodwaters inundating homes and buildings in southern Korea\n2. People evacuating flooded areas by boat\n3. Rescue workers helping those stranded by the floods\n4. Cars and buildings submerged in water\n5. Emergency responders distributing aid and supplies to affected communities",
"en_article_excerpt": "Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Korean_floods",
"event_category": "disasters",
"event_name": "2020_korean_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the cause and effects of the flooding, such as scenes of heavy rainfall, flooded areas, and rescue efforts. These events are likely to be visualized in a video documenting the floods and their impact on the region.\n\nEVENTS:\n1. Heavy rains pouring down in the southern parts of the Korean Peninsula\n2. Rivers overflowing and flooding nearby cities and towns\n3. People evacuating their homes and seeking higher ground\n4. Rescue teams navigating boats through flooded streets to reach stranded residents\n5. Cars and buildings submerged in water with debris scattered around",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the consequences of flooding and heavy rainfall, such as damage to infrastructure, evacuation efforts, and scenes of flooding. These events are likely to be visualized in a video capturing the impact of the floods on the region.\n\nEVENTS:\n1. People evacuating flooded homes and buildings\n2. Cars and buildings submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Debris and wreckage from destroyed buildings and infrastructure\n5. Residents clearing mud and water from their homes and streets",
"video_URL": "http://youtube.com/watch?v=nR08Eart4vo",
"video_description": "댐 관리를 담당하는 한국수자원공사가 이번 수해와 관련해 댐 수위 조절에는 문제가 없었다고 밝혔습니다. 수자원공사는 댐관리규정 등에 따라 홍수기 기상변화와 댐 상·하류 상황, 댐 안전과 민원 등을 고려해 홍수 대응이 가능하도록 댐 수위를 관리했다고 설명했습니다. 섬진강댐은 집중호우 전부터 홍수기제한수위보다 3m 낮게 유지했고, 용담댐이나 합천댐도 제한 수위를 지켜 전혀 이상이 없었다고 강조했습니다. 다만, 유례없는 최장 기간의 장마로 제방이 무너지고 물이 넘치는 등 복합적인 요인 탓에 홍수 피해가 난 것으로 보인다고 덧붙였습니다. 이에 관련 기관과 함께 원인 조사를 한 뒤 대책 수립에 적극 협조하고, 복구 지원에 노력하겠다고 전했습니다. ※ '당신의 제보가 뉴스가 됩니다' YTN은 여러분의 소중한 제보를 기다립니다. [카카오톡] YTN을 검색해 채널 추가 [전화] 02-398-8585 [메일] social@ytn.co.kr [온라인 제보] www.ytn.co.kr ▶ 기사 원문 : https://www.ytn.co.kr/_ln/0103_202008121704378843 ▶ 제보 안내 : http://goo.gl/gEvsAL, 모바일앱, social@ytn.co.kr, #2424 ▣ YTN 유튜브 채널 구독 : http://goo.gl/Ytb5SZ ⓒ YTN & YTN PLUS 무단 전재 및 재배포금지",
"video_id": "nR08Eart4vo"
}
| 4
|
[
"The frame shows an aerial view of a flooded area with extensive water coverage over fields and buildings, suggesting a severe flood. Text in Korean at the top reads \"내일까지 장마 소강,\" indicating a flood warning, and at the bottom, \"댐, 홍수기 제한수위 이하 유지...피해 원인 조사,\" mentioning that the dam and flood levels are below the restricted water level and there is an investigation into the cause of the damage.",
"The current frame shows an aerial view of a flooded area with extensive water coverage over fields and buildings, indicating severe flooding. Text in Korean at the top reads \"내일까지 장마 소강,\" suggesting a flood warning, and at the bottom, \"댐관리규정 등에 따라 홍수 대응 가능하도록 수위 관리,\" which translates to \"Water level management according to dam management regulations, etc., to enable flood response.\" The scene continues to depict the widespread flooding from the previous frame, with water covering large portions of the landscape, including roads and possibly residential areas. The text emphasizes the efforts to manage water levels in response to the flood, likely referencing actions taken by dam authorities to mitigate the effects of the flooding.",
"The current frame shows an aerial view of a flooded area with a river overflowing its banks, causing widespread water coverage over fields and roads, indicating severe flooding. Text in Korean at the top reads \"내일까지 장마 소강,\" suggesting a flood warning, and at the bottom, \"최장 시간 장마에 뒤덮이고 물 넘치는 복합 요인 탓,\" which translates to \"Due to the compound factors of the longest rainy season and overflowing water.\" The scene continues to depict the widespread flooding from the previous frame, with water covering large portions of the landscape, including roads and possibly residential areas. The text emphasizes the prolonged and severe nature of the flooding, likely referencing the extended rainy season and the contributing factors that have led to the current situation.",
"The current frame shows several individuals in high-visibility vests working amidst what appears to be wreckage or debris, possibly from a collapsed structure or an incident related to the ongoing severe flooding mentioned in the previous frame. The individuals are engaged in cleanup or assessment activities, suggesting efforts to manage the aftermath of the flooding. The text \"내일까지 장마 소강\" at the top of the image reiterates the flood warning and the prolonged nature of the rainy season, while \"최장 시간 장마에 뒤덮이고 물 넘치는 복합 요인 탓\" at the bottom emphasizes the compound factors leading to the flooding, including the prolonged rainy season and overflowing water. This frame depicts the human response to the flooding, indicating the beginning of recovery and assessment efforts."
] |
nR08Eart4vo
|
A severe flood has affected a region, causing widespread water coverage over fields, buildings, and roads. The flood warning, indicated by the text "내일까지 장마 소강," suggests that the flooding is expected to continue until the next day. The flooding is attributed to compound factors, including the longest rainy season and overflowing water, as stated in the text "최장 시간 장마에 뒤덮이고 물 넘치는 복합 요인 탓." Authorities are managing water levels according to dam management regulations to enable flood response. The video also shows the aftermath of the flooding, with individuals in high-visibility vests engaged in cleanup and assessment activities amidst wreckage or debris, marking the beginning of recovery efforts. Overall, the video depicts a severe flooding event and the initial response to manage and recover from the disaster.
|
Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.
|
[
"Heavy rains and floods in the southern Korean Peninsula during summer 2020",
"Heavy rains pouring down in southern South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North and South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during the summer floods of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during heavy rains in summer 2020",
"Rescue teams navigating boats through flooded streets to reach stranded residents during the summer floods of 2020 in North and South Korea",
"Rescue teams navigating boats through flooded streets to reach stranded residents in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after floods in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after heavy rains in North and South Korea during the summer of 2020"
] |
[
"People evacuating flooded homes and buildings in the Korean Peninsula during the summer rainy season of 2020",
"People evacuating flooded homes and buildings in the Korean Peninsula during summer 2020 heavy rains",
"Cars and buildings submerged in floodwater in North Korea and South Korea during the summer floods of 2020",
"Cars and buildings submerged in floodwater during heavy rains in the Korean Peninsula in summer 2020",
"Emergency responders rescuing people from flooded areas in North Korea and South Korea during the summer rainy season of 2020",
"Emergency responders rescuing people from flooded areas in North and South Korea during heavy rains in the summer of 2020",
"Debris and wreckage from destroyed buildings and infrastructure due to floods in North and South Korea during the summer 2020 rainy season",
"Debris and wreckage from destroyed buildings and infrastructure in North Korea, South Korea, and the Korean Peninsula after heavy rains during the summer rainy season of June-August 2020",
"Residents clearing mud and water from their homes and streets after floods in the Korean Peninsula during the summer 2020 rainy season",
"Residents clearing mud and water from their homes and streets in North and South Korea after heavy rains during the summer rainy season of 2020"
] |
[
"Floodwaters inundating homes and buildings in southern Korea, particularly in South Korea, during the summer rainy season of 2020",
"Floodwaters inundating homes and buildings in southern Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North and South Korea during heavy rains in summer 2020",
"Rescue workers helping those stranded by the floods in North Korea and South Korea during the summer rainy season of 2020",
"Rescue workers helping those stranded by floods in the Korean Peninsula during the summer rainy season of 2020",
"Cars and buildings submerged in water during floods in North and South Korea, Korean Peninsula, Summer 2020",
"Cars and buildings submerged in water in North and South Korea during heavy rains in summer 2020",
"Emergency responders distributing aid and supplies to communities affected by floods in North Korea and South Korea during the summer rainy season of June to August 2020",
"Emergency responders distributing aid and supplies to affected communities in North and South Korea during heavy rains in the summer of 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2020%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact of floods and heavy rains in Korea, such as scenes of flooding, damage, and rescue efforts, as these are likely to be visualized in a video about the event. The extracted events should be specific to the region and the time of year, taking into account the regional rainy season and the areas most affected by the floods.\n\nEVENTS:\n1. Floodwaters inundating homes and buildings in southern Korea\n2. People evacuating flooded areas by boat\n3. Rescue workers helping those stranded by the floods\n4. Cars and buildings submerged in water\n5. Emergency responders distributing aid and supplies to affected communities",
"en_article_excerpt": "Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Korean_floods",
"event_category": "disasters",
"event_name": "2020_korean_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the cause and effects of the flooding, such as scenes of heavy rainfall, flooded areas, and rescue efforts. These events are likely to be visualized in a video documenting the floods and their impact on the region.\n\nEVENTS:\n1. Heavy rains pouring down in the southern parts of the Korean Peninsula\n2. Rivers overflowing and flooding nearby cities and towns\n3. People evacuating their homes and seeking higher ground\n4. Rescue teams navigating boats through flooded streets to reach stranded residents\n5. Cars and buildings submerged in water with debris scattered around",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the consequences of flooding and heavy rainfall, such as damage to infrastructure, evacuation efforts, and scenes of flooding. These events are likely to be visualized in a video capturing the impact of the floods on the region.\n\nEVENTS:\n1. People evacuating flooded homes and buildings\n2. Cars and buildings submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Debris and wreckage from destroyed buildings and infrastructure\n5. Residents clearing mud and water from their homes and streets",
"video_URL": "http://youtube.com/watch?v=GwY9HYVEQIk",
"video_description": "전국 하천 홍수특보 전부 해제…상수도 93곳 피해 중부지방 집중호우가 소강상태에 접어든 가운데 전국 하천 주요 지점의 홍수특보가 전부 해제됐습니다. 환경부는 낙동강이 흐르는 경남 밀양시 삼랑진교에 내려졌던 홍수주의보가 오늘(12일) 오후 5시30분 해제된 것을 끝으로 전국 하천 주요지점에 내려진 홍수특보가 모두 해제됐다고 밝혔습니다. 한강과 임진강 비룡대교·영평교, 왕숙천 진관교 등지에 발령됐던 홍수주의보와 섬진강 하동군 읍내리 지점의 홍수주의보는 각각 지난 11일 오후 6시와 오후 5시 해제됐습니다. 이번 호우로 27개 시·군의 상수도 시설 93곳에서 피해가 발생했습니다. 연합뉴스TV 기사문의 및 제보 : 카톡/라인 jebo23 ▣ 연합뉴스TV 유튜브 채널 구독 https://goo.gl/VuCJMi ▣ 대한민국 뉴스의 시작 연합뉴스TV / Yonhap News TV http://www.yonhapnewstv.co.kr/",
"video_id": "GwY9HYVEQIk"
}
| 4
|
[
"The video frame shows a flooding scene under a concrete structure, likely a bridge, with muddy water flowing swiftly around the columns. Text in the image mentions flood-related news, indicating severe flooding and damage in an area, with references to the amount of rain and the impact on lives and property.",
"The current frame shows a bridge over a swollen river, with muddy water flowing swiftly around the columns. The text mentions a 'flooding emergency' and reports '93 houses' affected. The environment looks overcast, suggesting recent or ongoing heavy rainfall.",
"The current frame shows a flooded street with muddy water flowing along the road. The text mentions the 'flood damage in the area' and reports '93 houses' affected. The environment looks overcast, suggesting recent or ongoing heavy rainfall.",
"The current frame shows an indoor setting with flooded water covering the floor. The text mentions 'homes completely submerged' and reports 'all residents evacuated.' The environment looks dimly lit, suggesting the flooding might be causing power outages."
] |
GwY9HYVEQIk
|
A severe flooding event is depicted in the video, showing the impact of heavy rainfall on an area. The floodwaters are swift and muddy, affecting homes and infrastructure under a bridge and along streets, with 93 houses reported to be affected. The situation is described as a 'flooding emergency' with significant damage and disruption to lives and property. Residents have been evacuated from completely submerged homes, and the environment suggests recent or ongoing heavy rainfall, with possible power outages due to the flooding.
|
Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.
|
[
"Heavy rains and floods in the southern Korean Peninsula during summer 2020",
"Heavy rains pouring down in southern South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North and South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during the summer floods of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during heavy rains in summer 2020",
"Rescue teams navigating boats through flooded streets to reach stranded residents during the summer floods of 2020 in North and South Korea",
"Rescue teams navigating boats through flooded streets to reach stranded residents in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after floods in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after heavy rains in North and South Korea during the summer of 2020"
] |
[
"People evacuating flooded homes and buildings in the Korean Peninsula during the summer rainy season of 2020",
"People evacuating flooded homes and buildings in the Korean Peninsula during summer 2020 heavy rains",
"Cars and buildings submerged in floodwater in North Korea and South Korea during the summer floods of 2020",
"Cars and buildings submerged in floodwater during heavy rains in the Korean Peninsula in summer 2020",
"Emergency responders rescuing people from flooded areas in North Korea and South Korea during the summer rainy season of 2020",
"Emergency responders rescuing people from flooded areas in North and South Korea during heavy rains in the summer of 2020",
"Debris and wreckage from destroyed buildings and infrastructure due to floods in North and South Korea during the summer 2020 rainy season",
"Debris and wreckage from destroyed buildings and infrastructure in North Korea, South Korea, and the Korean Peninsula after heavy rains during the summer rainy season of June-August 2020",
"Residents clearing mud and water from their homes and streets after floods in the Korean Peninsula during the summer 2020 rainy season",
"Residents clearing mud and water from their homes and streets in North and South Korea after heavy rains during the summer rainy season of 2020"
] |
[
"Floodwaters inundating homes and buildings in southern Korea, particularly in South Korea, during the summer rainy season of 2020",
"Floodwaters inundating homes and buildings in southern Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North and South Korea during heavy rains in summer 2020",
"Rescue workers helping those stranded by the floods in North Korea and South Korea during the summer rainy season of 2020",
"Rescue workers helping those stranded by floods in the Korean Peninsula during the summer rainy season of 2020",
"Cars and buildings submerged in water during floods in North and South Korea, Korean Peninsula, Summer 2020",
"Cars and buildings submerged in water in North and South Korea during heavy rains in summer 2020",
"Emergency responders distributing aid and supplies to communities affected by floods in North Korea and South Korea during the summer rainy season of June to August 2020",
"Emergency responders distributing aid and supplies to affected communities in North and South Korea during heavy rains in the summer of 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2020%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact of floods and heavy rains in Korea, such as scenes of flooding, damage, and rescue efforts, as these are likely to be visualized in a video about the event. The extracted events should be specific to the region and the time of year, taking into account the regional rainy season and the areas most affected by the floods.\n\nEVENTS:\n1. Floodwaters inundating homes and buildings in southern Korea\n2. People evacuating flooded areas by boat\n3. Rescue workers helping those stranded by the floods\n4. Cars and buildings submerged in water\n5. Emergency responders distributing aid and supplies to affected communities",
"en_article_excerpt": "Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Korean_floods",
"event_category": "disasters",
"event_name": "2020_korean_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the cause and effects of the flooding, such as scenes of heavy rainfall, flooded areas, and rescue efforts. These events are likely to be visualized in a video documenting the floods and their impact on the region.\n\nEVENTS:\n1. Heavy rains pouring down in the southern parts of the Korean Peninsula\n2. Rivers overflowing and flooding nearby cities and towns\n3. People evacuating their homes and seeking higher ground\n4. Rescue teams navigating boats through flooded streets to reach stranded residents\n5. Cars and buildings submerged in water with debris scattered around",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the consequences of flooding and heavy rainfall, such as damage to infrastructure, evacuation efforts, and scenes of flooding. These events are likely to be visualized in a video capturing the impact of the floods on the region.\n\nEVENTS:\n1. People evacuating flooded homes and buildings\n2. Cars and buildings submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Debris and wreckage from destroyed buildings and infrastructure\n5. Residents clearing mud and water from their homes and streets",
"video_URL": "http://youtube.com/watch?v=FG82wp__uZ8",
"video_description": "#홍수 #성금 ●BTN불교TV 유튜브 구독하기 https://bit.ly/3k5nt10 (링크 누르고 구독 버튼 클릭!) ●BTN불교TV는? BTN불교TV(주)는 25년의 시간 동안 부처님의 법음을 널리 홍포해온 한국을 대표하는 불교방송국입니다. 불자님들의 소중한 후원금과 광고수입은 더 나은 방송 포교와 불교문화 진흥, 다양한 컨텐츠를 제작하는데 사용되고 있습니다. BTN은 앞으로도 방송 포교사의 소임을 다하겠습니다. ●BTN불교TV 채널번호 안내 IPTV: KTolleh:233번, SK B TV:305번, LGU+TV:275번(전국공통) SKYLIFE: 181번(전국공통) 지역케이블: 케이블 가입자의 경우 BTN채널번호가 지역별로 달라서 아래 주소를 클릭하셔서 확인하시면 됩니다 https://blog.naver.com/btnii/221391809765 ●다양한 경로로 BTN불교TV를 만나보세요! 홈페이지:http://www.btn.co.kr/index.asp 유튜브:https://www.youtube.com/channel/UCOozqNhXDcb6gg6oaU7uImA?view_as=subscriber 페이스북:https://www.facebook.com/BTNkorea/ 블로그:https://blog.naver.com/btnii 카카오스토리:https://story.kakao.com/ch/btntv",
"video_id": "FG82wp__uZ8"
}
| 4
|
[
"A news anchor in a pink blazer is reporting on a scene behind her, showing a traditional building with significant roof damage. The text at the bottom suggests a news headline about a fire-related incident.",
"The current frame depicts the interior of a traditional building, with the floor covered in water. In the background, there are several golden statues and a large painting on the wall, suggesting this is a place of worship or cultural significance. The text \"마당은 토사로 매워지고\" translates to \"The courtyard is being sealed with soil,\" indicating that efforts are being made to manage or contain the flooding situation. This frame continues the narrative of the previous scene by showing the extent of the damage and the ongoing efforts to address it.",
"The current frame shows several individuals actively engaged in cleaning up a muddy area outside a building. They appear to be shoveling and removing mud from the steps and around the entrance, possibly as part of a cleanup effort following a flood or similar incident. The text \"자비행을 실천해야 할 때입니다\" translates to \"It is time to practice compassion,\" suggesting a call for empathy and assistance in response to the situation. This frame continues the narrative by showing the community's response to the flooding damage.",
"The current frame depicts a silhouette of a person against a backdrop of a setting or rising sun, with bare tree branches visible. This image is overlaid with text in Korean that reads \"문의 1855-0108 수재민모금 827-01-0332-638 국민은행 (예금주: 불교텔레비전)\" which translates to \"Contact 1855-0108 Donation for flood victims 827-01-0332-638 Kookmin Bank (Depositor: Buddhist Television).\" The text suggests a call to action for donations to aid flood victims, continuing the theme of community response and assistance depicted in the previous frame."
] |
FG82wp__uZ8
|
A news report covers a fire-related incident at a traditional, culturally significant building, resulting in significant roof damage and flooding. The interior of the building, a place of worship, shows water-covered floors, golden statues, and a large painting, with efforts being made to manage the flooding by sealing the courtyard. The community responds to the damage with a cleanup effort, where individuals shovel and remove mud from the building's entrance and steps. The narrative concludes with a call to action, urging viewers to practice compassion and assist flood victims through donations, providing contact and banking information for those wishing to contribute.
|
Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.
|
[
"Heavy rains and floods in the southern Korean Peninsula during summer 2020",
"Heavy rains pouring down in southern South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North and South Korea during the summer of 2020",
"Rivers overflowing and flooding nearby cities and towns in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during the summer floods of 2020",
"People evacuating their homes and seeking higher ground in North and South Korea during heavy rains in summer 2020",
"Rescue teams navigating boats through flooded streets to reach stranded residents during the summer floods of 2020 in North and South Korea",
"Rescue teams navigating boats through flooded streets to reach stranded residents in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after floods in North and South Korea during the summer rainy season of 2020",
"Cars and buildings submerged in water with debris scattered around after heavy rains in North and South Korea during the summer of 2020"
] |
[
"People evacuating flooded homes and buildings in the Korean Peninsula during the summer rainy season of 2020",
"People evacuating flooded homes and buildings in the Korean Peninsula during summer 2020 heavy rains",
"Cars and buildings submerged in floodwater in North Korea and South Korea during the summer floods of 2020",
"Cars and buildings submerged in floodwater during heavy rains in the Korean Peninsula in summer 2020",
"Emergency responders rescuing people from flooded areas in North Korea and South Korea during the summer rainy season of 2020",
"Emergency responders rescuing people from flooded areas in North and South Korea during heavy rains in the summer of 2020",
"Debris and wreckage from destroyed buildings and infrastructure due to floods in North and South Korea during the summer 2020 rainy season",
"Debris and wreckage from destroyed buildings and infrastructure in North Korea, South Korea, and the Korean Peninsula after heavy rains during the summer rainy season of June-August 2020",
"Residents clearing mud and water from their homes and streets after floods in the Korean Peninsula during the summer 2020 rainy season",
"Residents clearing mud and water from their homes and streets in North and South Korea after heavy rains during the summer rainy season of 2020"
] |
[
"Floodwaters inundating homes and buildings in southern Korea, particularly in South Korea, during the summer rainy season of 2020",
"Floodwaters inundating homes and buildings in southern Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North Korea and South Korea during the summer rainy season of 2020",
"People evacuating flooded areas by boat in North and South Korea during heavy rains in summer 2020",
"Rescue workers helping those stranded by the floods in North Korea and South Korea during the summer rainy season of 2020",
"Rescue workers helping those stranded by floods in the Korean Peninsula during the summer rainy season of 2020",
"Cars and buildings submerged in water during floods in North and South Korea, Korean Peninsula, Summer 2020",
"Cars and buildings submerged in water in North and South Korea during heavy rains in summer 2020",
"Emergency responders distributing aid and supplies to communities affected by floods in North Korea and South Korea during the summer rainy season of June to August 2020",
"Emergency responders distributing aid and supplies to affected communities in North and South Korea during heavy rains in the summer of 2020"
] |
{
"article_url": "https://ko.wikipedia.org/wiki/2020%EB%85%84_%ED%95%9C%EA%B5%AD_%EC%A7%91%EC%A4%91%ED%98%B8%EC%9A%B0",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION: It makes sense to extract events that are directly related to the impact of floods and heavy rains in Korea, such as scenes of flooding, damage, and rescue efforts, as these are likely to be visualized in a video about the event. The extracted events should be specific to the region and the time of year, taking into account the regional rainy season and the areas most affected by the floods.\n\nEVENTS:\n1. Floodwaters inundating homes and buildings in southern Korea\n2. People evacuating flooded areas by boat\n3. Rescue workers helping those stranded by the floods\n4. Cars and buildings submerged in water\n5. Emergency responders distributing aid and supplies to affected communities",
"en_article_excerpt": "Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.",
"en_article_url": "https://en.wikipedia.org/wiki/2020_Korean_floods",
"event_category": "disasters",
"event_name": "2020_korean_floods",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the cause and effects of the flooding, such as scenes of heavy rainfall, flooded areas, and rescue efforts. These events are likely to be visualized in a video documenting the floods and their impact on the region.\n\nEVENTS:\n1. Heavy rains pouring down in the southern parts of the Korean Peninsula\n2. Rivers overflowing and flooding nearby cities and towns\n3. People evacuating their homes and seeking higher ground\n4. Rescue teams navigating boats through flooded streets to reach stranded residents\n5. Cars and buildings submerged in water with debris scattered around",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 26 Jul 2024\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: Between June and August 2020, floods severely impacted large areas of both North and South Korea due to heavy rains of the regional rainy season, primarily in the far southern parts of the Korean Peninsula.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION: Given the query about floods in North and South Korea due to heavy rains, it makes sense to extract events that are directly related to the consequences of flooding and heavy rainfall, such as damage to infrastructure, evacuation efforts, and scenes of flooding. These events are likely to be visualized in a video capturing the impact of the floods on the region.\n\nEVENTS:\n1. People evacuating flooded homes and buildings\n2. Cars and buildings submerged in floodwater\n3. Emergency responders rescuing people from flooded areas\n4. Debris and wreckage from destroyed buildings and infrastructure\n5. Residents clearing mud and water from their homes and streets",
"video_URL": "http://twitter.com/i/status/1291976724456992776",
"video_description": "섬진강이 난리네요 ㅠ#섬진강#유등면#섬진강 홍수",
"video_id": "1291976724456992776"
}
| 4
|
[
"The image shows a flooded area with muddy water covering the ground. There is a large blue metal structure, possibly a warehouse or shelter, partially submerged. A tree and some greenery are visible on the right side.",
"The image continues to depict a flooded area with muddy water covering the ground. The large blue metal structure remains partially submerged, and the tree and some greenery are still visible on the right side.",
"The scene remains consistent with the previous frame, showing the floodwaters continuing to rise around the large blue metal structure. The muddy water covers more of the area, including the space under the structure, and the greenery on the right side is still partially visible, indicating the ongoing impact of the flooding. The animals are still visible and affected by the flood, suggesting the floodwaters are deepening and spreading.",
"The floodwaters have risen significantly, further covering the area under the large blue metal structure. The greenery on the right side is mostly submerged, indicating the flood's progression. The animals are more submerged, and the water appears to be deepening and spreading."
] |
1291976724456992776
|
A severe flood is depicted, showing a large area covered in muddy water, with a partially submerged blue metal structure, possibly a warehouse or shelter. The floodwaters continuously rise, engulfing more of the surrounding area, including the space under the structure and the nearby greenery. The scene suggests a natural disaster, likely caused by heavy rainfall or a nearby body of water overflowing, with the flood's impact worsening over time, affecting the environment and potentially local wildlife.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.