query stringclasses 259
values | prequel listlengths 0 30 | sequel listlengths 0 30 | during listlengths 0 30 | metadata dict | num_of_frames int64 32 32 | frame_captions listlengths 32 32 | video_id stringlengths 6 45 | frame2video_caption stringlengths 287 3.07k | asr dict |
|---|---|---|---|---|---|---|---|---|---|
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... | [
"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 rubb... | [
"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 ea... | {
"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"
} | 32 | [
"The image shows a bustling street market in an urban setting at night, with people gathered around various stalls. The text on the image reads \"어제 저녁 7시 44분,\" which translates to \"yesterday evening at 7:44 PM.\" The scene depicts individuals walking, standing, and engaging in activities typical of a market, suc... | t8lY-rjy7-Y | The video depicts a series of scenes related to earthquake preparedness and response in South Korea. It begins with a bustling street market at night, followed by scenes of students participating in an earthquake drill in a classroom and a hallway. The drill appears to be part of a larger emergency preparedness exercise, as indicated by the involvement of various settings, including a train station and a storage room.
As the video progresses, it shifts to show the aftermath of an earthquake in Gyeongju, South Korea, with scenes of debris, destruction, and emergency responders. The footage includes news reports and official announcements, highlighting the severity of the earthquake and the government's response.
The latter part of the video focuses on governmental meetings and discussions, where officials address the earthquake situation, emphasizing the need for preparedness and learning from past experiences to better respond to future seismic events. The discussions also touch on criticism of the government's response to the earthquake, with calls for improved preparedness and emergency measures.
Throughout the video, the emphasis is on earthquake preparedness, response, and the importance of learning from past experiences to enhance safety measures for the future. The various scenes, from drills to governmental discussions, underscore the multifaceted approach to addressing natural disasters in South Korea. | {
"original": " I hope that the earthquake will be prepared for a bigger one in the future.",
"refined": "I hope that we will be better prepared for a bigger earthquake in the future.",
"summarized": "The speaker expresses a hope that preparations will be in place for a potentially larger earthquake in the future, and highlights the importance of predicting and preparing for great disasters, possibly referring to a specific event expected in 5 years.",
"translated": "I hope that we will be prepared for a bigger earthquake in the future.",
"translated_llm": "I hope that the earthquake will be prepared for a bigger one in the future.",
"translated_whisper": " The The The The The The The The The The The The The The The The The The The The The The The The The The The The The ... ... ... ... ... ... ... ... ... ... ... ... ... ... Even the most difficult thing is A huge disaster The disaster is coming And now we need more Great disasters to be predicted A new emergency in 5 years That day Now This place will be"
} |
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... | [
"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 rubb... | [
"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 ea... | {
"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"
} | 32 | [
"In the current frame, a person is leaning over a desk in an office setting, seemingly focused on analyzing or reporting information on a computer. The visible text in the image includes \"기상청 사라남들\" which translates to \"Meteorological Office People\" and \"예고없이 찾아오는 지진 365일 근정하는 사람들편\" which translates to \"Peopl... | CXe6_CDrdB8 | The video depicts the work of individuals involved in earthquake monitoring and analysis, highlighting the importance of swift and accurate action in responding to seismic events. The setting is primarily an office or control room environment, where personnel are seen analyzing data on computer screens, discussing their findings, and collaborating on earthquake reports. The video focuses on a significant earthquake that occurred in Gyeongju, which was initially reported as a magnitude 1.1 but later determined to be much larger, at 5.8. The individuals involved express a sense of anxiety and tension due to the unpredictability of earthquakes and the critical nature of their work. Despite these challenges, they demonstrate a professional demeanor and a commitment to providing accurate and timely information. The video also touches on the use of early warning systems and the importance of quick response in earthquake management. Throughout, the atmosphere remains tense and focused, reflecting the high stakes and critical nature of earthquake analysis and response. The personnel involved take pride in their work, acknowledging the challenges and responsibilities that come with being in charge of earthquake duty, and they emphasize the need for precision and urgency in managing seismic data. | {
"original": " 국내 뿐만 아니라 국외에서 발생하는 지진을 분석을 해서 일정 규모 이상이 되면 통보를 하고 있고요. 더불어서 화산 정보를 제공을 하고 있습니다. 경제에서 발생한 지진인데요. 아찔한데요. 당시만 해도 규모가 큰 지진은 별로 없었거든요. 분석하고 있는 분은 엄청 떨리고 식은땀은 나고 일단 불안감이 먼저 앞섰습니다. 1.1의 지진을 통보를 했는데 40분, 50분 뒤에 앞서 발표한 지진보다 훨씬 큰 게 감이 오더라고요. 최대한 더 빨리 해야 되겠다 분석을 시작했는데 규모 5.8이 나왔어요. 충분히 만족을 하고 있습니다. 지진이 어느 시각에 발생할지 모르기 때문에 항상 불안하고 긴장감을 가지고 근무하고 있지만 지진을 맡고 있는 담당자의 숙명이라고 생각합니다. 지진이 배우님, 지진 업무를 담당하고 있는 사람으로서 고마움을 느끼고 있습니다. 화이팅입니다. 지진, 올바로 알아야 빠르게 대처할 수 있습니다. 지진은 신속정학입니다. 제가 그 신속정학의 중심에 서겠습니다",
"refined": "In Korea and abroad, we analyze earthquakes and report those that exceed a certain magnitude. Additionally, we provide volcanic information. I'm referring to the earthquake that occurred in the economy. It was quite alarming. At the time, there weren't many large earthquakes, so the analyst was extremely nervous and anxious. After reporting a 1.1 magnitude earthquake, we felt that a much larger one was coming 40 to 50 minutes later. We began analyzing as quickly as possible and determined it was a 5.8 magnitude earthquake. I'm satisfied with the results. Since earthquakes can occur at any time, I'm always anxious and on edge, but I consider it an occupational hazard. As someone in charge of earthquake duty, I feel grateful. I'll do my best to provide accurate information, as understanding earthquakes correctly is crucial for prompt action.",
"summarized": "The speaker is an earthquake analyst who monitors and reports earthquakes in Korea and other countries. They recall a particular earthquake in Gyeongju that initially seemed small, but later turned out to be much larger, with a magnitude of 5.8. The analyst expresses satisfaction with their work and acknowledges the anxiety and tension that comes with the job, but feels a sense of responsibility and gratitude for being able to contribute to public safety through their work. The speaker emphasizes the importance of accurate and rapid information about earthquakes to facilitate quick response and preparedness.",
"translated": "We analyze earthquakes not only in Korea but also in other countries, and if they exceed a certain magnitude, we report them. Additionally, we provide volcanic information. This is about an earthquake that occurred in the economic sector, which was quite shocking. At the time, there weren't many large earthquakes, so the person analyzing it was extremely nervous, sweating, and felt a great deal of anxiety. We reported an earthquake with a magnitude of 1.1, but 40 to 50 minutes later, it became clear that it was going to be much larger than initially announced. We began analyzing as quickly as possible and determined that the magnitude was 5.8. I'm satisfied with the result. Since we can't predict when an earthquake will occur, we always work with a sense of anxiety and tension, but I believe this is the nature of being in charge of earthquake duty. As someone responsible for earthquake work, I feel grateful. Earthquakes require swift and accurate action, so it's essential to understand them correctly. I will be at the forefront of this rapid response.",
"translated_llm": "The earthquake is a major economic event. It is a painful event. It was not a major earthquake at the time. The person who was analyzing it was very shaken and sweating and I was very anxious. I reported a 1.1 earthquake, but after 40 minutes, 50 minutes, I was told that it was going to be much bigger than the one I had announced earlier. I started the analysis as soon as possible and it was 5.8. I am satisfied.",
"translated_whisper": " What is the earthquake in Korea? We analyze earthquakes in Korea and other countries and if they are of a certain size, we report them. In addition, we provide damage information. This is an earthquake from Gyeongju. It's a bit dizzy. At that time, there weren't many large earthquakes. I was very nervous and cold when I was analyzing. I was worried at first. I reported an earthquake at 1.2. After 40 to 50 minutes, I felt that it was much bigger than the previous earthquake. I started analyzing and I got a 5.8 scale. Q. How do you feel about the earthquake? I'm satisfied with the results. I don't know what the earthquake will look like, so I'm always anxious and nervous. But I think it's the life of the person in charge of the earthquake. Q. How do you feel about the earthquake? As an earthquake actor, I feel grateful as a person in charge of the earthquake. I'm cheering for you. Q. What is the earthquake? You have to know the earthquake correctly to react quickly. The earthquake is a rapid earthquake. I will be the center of that rapid development."
} |
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... | [
"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 rubb... | [
"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 ea... | {
"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"
} | 32 | [
"In the current frame, two news anchors are presenting a news segment. The male anchor on the left is dressed in a suit with a tie, while the female anchor on the right is wearing a blazer with a white blouse. The background shows a newsroom setting with multiple monitors displaying various images and graphics. The... | uF3x63vtddc | The video discusses the potential for predicting earthquakes through changes in groundwater levels, focusing on the recent earthquakes in Gyeongju, South Korea. Two news anchors present a news segment on the topic, citing the work of Professor Jeong Sang-yong, who analyzed data from an underground water observation station near the epicenter of the earthquakes. The data showed significant rises in groundwater levels a day or two before the earthquakes, suggesting a possible correlation between groundwater level changes and earthquake prediction. The video includes footage of residential areas, surveillance camera views, and people working on rooftops, possibly repairing damage from the earthquakes. The overall context is one of scientific investigation and discussion about the potential for early warning systems for earthquakes based on groundwater level changes. The video aims to inform viewers about the research and its implications for earthquake prediction and preparedness. | {
"original": " 지난 11일과 19일 경주에서 규모 5.8을 비롯한 대형 지진이 잇따라 발생하기 하루 이틀 전, 주변 지하수의 수위가 크게 상승해 향후 지진 예측에 활용할 수 있다는 주장이 나왔습니다. 정상용 부경대 지구환경과학과 교수는 진앙에서 가장 가까운 경북 경주시 산내면에 있는 지하수 관측소 자료를 분석한 결과 이같이 밝혔습니다. 자료를 보면 지난 11일 안반 지하수 수위는 하루 사이에 무려 49cm가 상승했고, 17일에도 39cm나 올라갔습니다. 정 교수는 지진에 앞서 암석에 강한 압력이 가해져 지하수의 수위가 많이 올라간 것으로 보인다며 규모 4.5 이상의 지진에만 영향을 받았다고 설명했습니다.",
"refined": "On the 11th and 19th, large earthquakes, including a 5.8-magnitude quake, occurred in Gyeongju. A day or two before the earthquakes, the groundwater level in the surrounding area rose significantly, which could be useful for predicting future earthquakes. Professor Jung of the Department of Earth and Environmental Science at Pukyong National University analyzed data from a groundwater observation facility in Sanin-myeon, Gyeongju, closest to the epicenter. The data shows that the groundwater level in Anban rose by 49cm in one day on the 11th and by 39cm on the 17th. According to Professor Jung, strong pressure on the rock prior to the earthquake caused a significant rise in the groundwater level, which was only affected by earthquakes with a magnitude of 4.5 or higher.",
"summarized": "Research by Professor Jung of the Earth Environment Science department at Pukyong National University suggests that changes in underground water levels may be used to predict future earthquakes. Analyzing data from an underground water observation facility in Gyeongju, the professor found that the water level rose significantly, 49cm on the 11th and 39cm on the 17th, before recent large earthquakes in the area. According to Professor Jung, the increased water level is likely due to strong pressure on rocks before an earthquake, and this phenomenon is only observed with earthquakes of magnitude 4.5 or higher.",
"translated": "A day or two before the recent large earthquakes, including a 5.8-magnitude quake, struck Gyeongju on the 11th and 19th, the groundwater level in the surrounding area rose significantly, prompting claims that this phenomenon could be used to predict future earthquakes. Professor Jeong Sang-yong of the Department of Earth and Environmental Science at Pukyong National University analyzed data from an underground water observation station in Sanneom-myeon, Gyeongju, which is closest to the epicenter, and reported these findings. According to the data, the groundwater level in Anban rose by 49cm in just one day on the 11th and by 39cm on the 17th. Professor Jeong explained that the increase in groundwater level appeared to be caused by strong pressure on the rocks before the earthquake and that only earthquakes with a magnitude of 4.5 or higher had an impact.",
"translated_llm": "The data from the nearest underground water observatory in the mountains of North Korea, located in the heart of North Korea, showed that the underground water level rose by 49 cm in the past 11 days and rose by 39 cm on the 17th. The professor explained that the underground water level appeared to have risen a lot before the earthquake and that the earthquake was only affected by a strong pressure on the rocks.",
"translated_whisper": " On the 11th and 19th, a large earthquake of 5.8 in Gyeongju, two days before the earthquake, The number of underground water around the area has risen significantly, and it is estimated that it can be used to predict the future. Professor of Earth Environment Science at the National University of Gyeongju, Gyeongsang-do, The data of the underground water observation facility in the mountains in Gyeongju, which is the closest to the Jinhang, has been analyzed. According to the data, the underground water level in Anban was 49cm higher in a day on the 11th, and it rose to 39cm on the 17th. Professor Jung said that the strong pressure of the rock was applied to the earthquake, and the underground water level rose a lot, and that it was only affected by an earthquake of 4.5 or more."
} |
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... | [
"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 rubb... | [
"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 ea... | {
"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"
} | 32 | [
"The current frame shows two news anchors seated at a desk in a news studio. The male anchor is on the left wearing a dark suit and red tie, while the female anchor is on the right dressed in a blue blazer and light pink blouse. Behind them, there is a split-screen image: the left side shows a nighttime street scen... | nln3QFJggKs | The video depicts the aftermath of an earthquake in Gyeongju, South Korea, focusing on the assessment and revision of the official damage estimate. Initially expected to exceed 10 billion won, the estimate was revised downward to approximately 5.8 billion won following a thorough survey. The revision was due to the exclusion of private property damages such as cracks in walls, broken fences, and decreased building foundations, which are not considered official property damage according to regulations. Only cases of complete or partial destruction of homes were included in the official tally. The video shows news anchors discussing the revised estimate, aerial views of the city, and scenes of damaged buildings, roads, and infrastructure, highlighting the extent of the earthquake's impact. Workers are seen engaged in repair activities, such as replacing broken tiles on traditional Korean roofs, and individuals are inspecting damaged houses and public areas, including cracked sidewalks and roads. The video emphasizes the challenges in assessing and addressing the earthquake's impact, particularly in distinguishing between included and excluded types of damage in the official damage estimate. Overall, the video provides a detailed look at the earthquake's aftermath in Gyeongju, the revision of the damage estimate, and the ongoing efforts to repair and recover from the disaster. | {
"original": " 이딴 강진으로 100억 원이 넘는 것으로 추정됐던 경주시의 공식 피해 규모가 58억 원 수준에 그치는 것으로 집계됐습니다. 경주시가 지진 발생 뒤 전수조사를 통해 발표한 잠정추산 피해액은 약 107억 원이었지만 사유재산 피해는 대부분 공식 피해액에서 빠졌습니다. 규정상 기화가 떨어지거나 벽체에 금이 가고 담장이 파손된 것은 재산 피해로 산정되지 않습니다. 사유재산 중에는 지진으로 주택이 완전히 부서지거나 반쯤 허물어진 것만 공식 피해에게 들어갔습니다.",
"refined": "The official damage in Gyeongju, initially estimated to exceed 10 billion won, has been recalculated to around 5.8 billion won. Following a thorough investigation after the earthquake, the city reported a preliminary estimated damage of approximately 10.7 billion won, but most private property damage was excluded from the official total. According to regulations, damage such as cracks in walls or broken fences is not considered property damage. Only private properties that were completely destroyed or partially demolished due to the earthquake were included in the official damage report.",
"summarized": "Gyeongju city's official damage from the earthquake was estimated to be over 1 billion won, but the final calculation is around 580 million won. The initial estimate of 1.07 billion won was reduced as most private property damages, such as cracks in walls or broken fences, were not included due to regulations. Only severe cases of private property damage, such as completely or partially destroyed houses, were accounted for in the official damage total.",
"translated": "The official damage estimate for Gyeongju, initially expected to exceed 10 billion won, has been revised to approximately 5.8 billion won. Following a thorough survey after the earthquake, the city reported preliminary estimated damages of around 10.7 billion won. However, most private property damages were not included in the official tally. According to regulations, damages such as cracks in walls, broken fences, or decreased building foundations are not considered property damage. Only cases where homes were completely or partially destroyed due to the earthquake were counted in the official damage estimate.",
"translated_llm": "The official damage estimated by the city of 경주 to be over 10 billion won was estimated to be at 58 billion won. The estimated damage was about 107 billion won, but most of the damage was lost from the official damage.",
"translated_whisper": " The official damage of Gyeongju, which was estimated to be over 1 billion won, has reached the level of 5.8 billion won. The total damage of the dam was about 1.7 billion won, which was announced through a full-scale survey after the earthquake. But most of the damage to the property was removed from the official damage. According to the regulations, the damage to the dam is not determined as damage to the property. Accordingly, the property was damaged by an earthquake, or half of the property was destroyed."
} |
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... | [
"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 rubb... | [
"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 ea... | {
"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"
} | 32 | [
"The current frame shows two news anchors presenting a report on an earthquake and its aftershocks. The anchor on the left is dressed in traditional Korean attire and appears to be reading from a script, while the anchor on the right, also in traditional attire, stands beside a map displaying seismic activity. The ... | n6UmQ9Kw2GQ | The video depicts the aftermath of a 5.8 magnitude earthquake that struck Gyeongju, focusing on the news coverage and the effects of the earthquake on daily life. The news anchors, dressed in traditional Korean attire, report on the earthquake and its 307 aftershocks, with the majority ranging from 1.5 to 3.0 in magnitude. The anchors stand in front of a cityscape background and beside a map displaying seismic activity, highlighting the location of the earthquake and its aftershocks.
The video transitions to CCTV footage of various indoor settings, including a hallway, a kitchen, and a living room, showing no significant damage or disturbance, emphasizing the resilience and normalcy of daily life despite the aftershocks. The footage also shows everyday objects, such as a baby bouncer and bottles of water, appearing undamaged and intact.
Throughout the video, the on-screen text reinforces the message that the intensity of the aftershocks is diminishing, with future aftershocks expected to have a magnitude of 3.0 or less, according to the Korea Meteorological Administration. The overall narrative suggests a gradual return to normalcy following the earthquake, with the video providing a coherent and reassuring account of the event and its aftermath. | {
"original": " 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 1.5에서 3.0도 292회 등 모두 307차례 발생했습니다. 기상청 관계자는 경주 지진 탓에 여진이 계속 발생하고 있지만 강도가 약해지고 있다며 앞으로 일어날 여진 규모는 3.0 이하로 예상되기 때문에 그렇게 우려하지 않아도 될 것 같다고 말했습니다.",
"refined": "On the evening of the 12th, a 5.8 magnitude earthquake occurred in Gyeongju, resulting in 307 aftershocks. According to the weather agency, as of 4 pm today, there have been 307 aftershocks with magnitudes ranging from 1.5 to 3.0 since the Gyeongju earthquake. An agency official stated that although aftershocks are continuing due to the Gyeongju earthquake, their intensity is weakening, and it is expected that future aftershocks will have a magnitude of 3.0 or less, so there is no need to worry.",
"summarized": "A magnitude 5.8 earthquake occurred in Gyeongju on the 12th, resulting in 307 aftershocks, with 292 of them ranging from 1.5 to 3.0 in magnitude. According to the Korea Meteorological Administration, although aftershocks are still occurring due to the earthquake, their intensity is weakening. Officials expect future aftershocks to have a magnitude of 3.0 or less, indicating no significant concern is warranted.",
"translated": "Following the 5.8 magnitude earthquake that struck Gyeongju on the evening of the 12th, a total of 307 aftershocks have been recorded. According to the Korea Meteorological Administration, as of 4 pm today, there have been 307 aftershocks, with 292 of them ranging from 1.5 to 3.0 in magnitude. An official from the Meteorological Administration noted that while aftershocks are still occurring due to the Gyeongju earthquake, their intensity is weakening, and it is expected that future aftershocks will have a magnitude of 3.0 or less, so there is no need for significant concern.",
"translated_llm": "The latest estimates are that 307 earthquakes have occurred due to the 5.8 magnitude earthquake that occurred on the 12th evening. According to the Meteorological Office, as of 4 pm today, the earthquake has caused 307 earthquakes ranging from 1.5 to 3.0 degrees and 292 rounds.",
"translated_whisper": " The last 12 days, there were 5.8 times of earthquakes in Gyeongju, and 307 times of the earthquake in Yeojin. According to the Ministry of Emergency Situations, Yeojin has been 3 times in Gyeongju since the earthquake in Gyeongju, from 1.5 to 3.0 degrees 292 times. The Kisangcheon official said that the earthquake is still happening due to the Gyeongju earthquake, but the intensity is getting weaker, and the expected earthquake size to occur in the future is 3.0 or less, so it is not necessary to worry."
} |
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... | [
"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 rubb... | [
"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 ea... | {
"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"
} | 32 | [
"The current frame shows a news anchor standing in front of a large screen displaying a surveillance camera image. The image on the screen depicts a dark outdoor scene, possibly affected by the recent earthquakes in the Gyeongju area. The news anchor, holding a white folder, appears to be delivering a report on the... | uoagwCVoHGM | A news report covers the aftermath of an earthquake in the Gyeongju area, which resulted in 8 injuries and 253 cases of property damage. The earthquake caused disruptions to communication networks, including KakaoTalk, and temporarily halted the KTX train on the Daejeon-Busan route. The report shows surveillance footage of damaged buildings, broken pipes, and debris-filled streets, as well as people cleaning up and assessing the damage. The news anchor provides updates on the situation, including the number of injuries and property damage, and reports on the efforts to restore order and normalcy in the affected areas. The video also shows scenes of daily life returning to normal, with people traveling through airports and subway stations, highlighting the resilience of the community in the face of natural disasters. | {
"original": " 어제저녁 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 재산 피해가 신고됐다고 밝혔습니다. 이번 지진으로 인해 KTX 대전에서 부산 구간이 5분가량 정차하기도 했습니다. MBC 뉴스 김성현입니다.",
"refined": "Yesterday evening, an earthquake in Gyeongju left 8 people injured and caused temporary disruptions to communication networks, including Kakao Talk and the National Safety Agency's website. The National Safety Agency reported that the earthquake, which occurred in Gyeongju over the past two days, resulted in 8 people, including an 80-year-old woman, sustaining minor injuries, and 253 cases of property damage. The earthquake also caused the KTX train service between Daejeon and Busan to be suspended for about 5 minutes.",
"summarized": "An earthquake in Gyeongju resulted in 8 people being injured, including an 80-year-old woman, and 253 cases of property damage reported. The incident also caused a temporary disruption to communication networks, including Kakao Talk and the National Security Agency website, and a 5-minute halt to the KTX train service between Daejeon and Busan.",
"translated": "Yesterday evening, an earthquake in the Gyeongju area resulted in 8 injuries and temporarily disrupted communication networks, including KakaoTalk and the National Security Agency's website. The National Security Agency reported that 8 people, including an 80-year-old woman named Hwang, were injured and 253 cases of property damage were reported due to the earthquakes that occurred in Gyeongju yesterday and today. The earthquake also caused the KTX train to halt for about 5 minutes on the Daejeon-Busan route. This is MBC News reporter Kim Seong-hyeon.",
"translated_llm": "The earthquake in the raceway yesterday evening caused 8 people to be injured, and the telecommunications network including the KaKaoTok and the website of the National Security Agency were temporarily disrupted. The National Security Agency said that eight people including an 80-year-old woman, Mr. Zhuang Mo, were injured and 253 property damage was reported due to the earthquake in the raceway today and yesterday.",
"translated_whisper": " Yesterday evening, 8 people were injured in the earthquake in Gyeongju, and there was a problem with the communication network such as Kakao Talk and the National Security Agency website. The National Security Agency reported that 8 people, including an 80-year-old woman, were injured in the earthquake that occurred yesterday and today in Gyeongju. 253 property damage was reported. Due to the earthquake, the Busan section of KTX Daejeon was also closed for about 5 minutes."
} |
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... | [
"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 rubb... | [
"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 ea... | {
"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"
} | 32 | [
"The current frame shows two news anchors, a man and a woman, dressed in traditional Korean attire, presenting a news segment. The background features a panoramic view of a cityscape with a river and a bridge. The text on the screen indicates the news channel \"연합뉴스TV\" and the time \"15:01.\" The context of the fr... | 5p1N4MtudiI | The video appears to be a news broadcast discussing the aftermath of a 5.8-magnitude earthquake that occurred in Gyeongju, South Korea, 12 days prior to the broadcast. The earthquake was followed by 316 aftershocks, with the majority (301) ranging in magnitude from 1.5 to 3.0, 14 ranging from 3.0 to 4.0, and one ranging from 4.0 to 5.0. The news anchors, dressed in traditional Korean attire, present the information in front of a panoramic view of a cityscape with a river and a bridge. The broadcast includes updates from the Korea Meteorological Administration, which reports that the aftershocks have been subsiding. The video also shows maps and data related to the seismic activity, including the locations and magnitudes of the aftershocks, as well as a digital counter displaying the number of aftershocks. The news broadcast aims to inform viewers about the ongoing seismic activity in Gyeongju and provide updates on the situation. | {
"original": " 지난 12일에 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.",
"refined": "The aftershocks of the 5.8 magnitude Gyeongju earthquake that occurred on the 12th have been recorded 316 times. As of 11 am this morning, the Korea Meteorological Administration reported 316 aftershocks. The breakdown by magnitude is as follows: 301 aftershocks between 1.5 and 3.0, 14 between 3.0 and 4.0, and one between 4.0 and 5.0. The aftershocks, which have continued for three days, are showing signs of subsiding, according to the Korea Meteorological Administration.",
"summarized": "A 5.8-magnitude earthquake occurred in Gyeongju on the 12th, followed by 316 aftershocks as of 11 am today. The aftershocks were mostly minor, with 301 between 1.5 and 3.0 magnitude, 14 between 3.0 and 4.0, and 1 between 4.0 and 5.0. The aftershocks have continued for four days but appear to be subsiding, according to the Korea Meteorological Administration.",
"translated": "The aftershocks of the 5.8-magnitude earthquake that occurred in Gyeongju 12 days ago have been recorded 316 times. As of 11 am today, the Korea Meteorological Administration reported that the number of aftershocks has reached 316. By magnitude, 301 of the aftershocks ranged from 1.5 to 3.0, 14 ranged from 3.0 to 4.0, and one ranged from 4.0 to 5.0. The aftershocks, which have continued for three days, are showing signs of subsiding, according to the Korea Meteorological Administration.",
"translated_llm": "The number of earthquakes in the last 12 days has been estimated at 316 of magnitude 5.8. The Meteorological Office has announced that the number of earthquakes as of 11 am today is 316. The number of earthquakes in the past 12 days has been estimated at 1.5 to 3.0 and 301 times, 3.0 to 4.0 and 14 times, 4.0 to 5.0 and once.",
"translated_whisper": " The incident occurred on 5.8 Gyeongju. Three hundred and sixteen times of earthquake occurred. As a result. The Kisangcheon News Agency reported that the number of earthquakes occurred at 11 a.m. this morning as 316. The number of a 1.5 to 3.0 is 301 times. Three three points zero four points zero were a few zeroes, four points zero and a quarter is a few. I have been detected. Over the past four days, the 여진 has been seen to be stuck for a while. Kisang-chung is knowing that."
} |
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... | [
"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 rubb... | [
"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 ea... | {
"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"
} | 32 | [
"The frame depicts a traditional Korean neighborhood with a dense collection of buildings featuring distinctive black-tiled roofs. The visible text reads \"하늘에서 본 경주 지진피해 복구 현장,\" which translates to \"Aerial view of Gyeongju earthquake damage recovery site.\" The image suggests that the video is showing the recons... | roaxxb8zRLU | The video depicts the aftermath and recovery efforts following an earthquake in Gyeongju, South Korea. The footage shows aerial views of traditional Korean neighborhoods with densely packed buildings featuring black-tiled roofs, highlighting the extent of the damage and the ongoing reconstruction efforts. The scenes focus on the architectural style and the recovery activities in Hwanam-dong, Gyeongju, with temporary measures such as tents and tarpaulins indicating community services and protection for the structures.
The video also showcases the collaborative effort in rebuilding, with restaurants participating in recovery efforts and workers engaged in repair and reconstruction activities. The aerial views emphasize the scale and density of the residential area, underscoring the community's resilience and the challenges faced in the recovery process.
Additionally, the footage features historical and cultural sites in Gyeongju, such as Cheomseongdae Observatory, which appears less crowded than usual. The scenes provide a unique perspective on the site's typical popularity and the impact of the earthquake on local tourism.
The video also covers the earthquake's epicenter and its proximity to a school and a rural settlement, highlighting the potential disruption to transportation and daily life. Aerial views of the surrounding rural area show the extent of the earthquake's impact, with red overlays indicating potential transportation disruptions and maps detailing seismic activity.
The final frames feature the logo of Kookje Shinmun, a news or media organization, suggesting their involvement in reporting or covering the earthquake event. Overall, the video provides a comprehensive view of the earthquake's aftermath and the community's efforts to recover and rebuild. | {
"original": " ល្ពបានដោយើងមំខងានដំល ១ េការអណាញុណុណាស្នៃទីពះម្អាចំង ១ងះបំ្រឿ្រភង្តាថាពវច្ញុ់វាសដំង ៩ចម្រាប់ប់ព្បចេតដែ និងយបាចំទេណល្នៅមិនោយើងរឹកខ្តើពយត។ សល អាចប្រាប់ប់ម្ន... លៈក សូតប្រឿស សុវីបាន់បញុងថ្ពេក ទានយ។ កំពុជ។ ឯកំពុជ្តលើយៈ។ ឡើលយងឬហើយចំផុជោយ ចុង។",
"refined": "The Khmer script has been written by us for about a thousand years, and we have kept it until now. If we lose Khmer script, it means we lose Khmer identity.",
"summarized": "Not Available",
"translated": "Not Available",
"translated_llm": "The first one is a very strong one, and it's a very strong one, and it's a very strong one, and it's a very strong one, and it's a very strong one. It's a very strong one, and it's a very strong one.",
"translated_whisper": " The"
} |
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... | [
"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 rubb... | [
"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 ea... | {
"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"
} | 32 | [
"The current frame shows a news broadcast discussing President Park Geun-hye's recent instructions regarding the declaration of a special disaster area for the Gyeongju region in Gyeongbuk, following a series of strong earthquakes. In the background, a group of people, some wearing yellow jackets, are gathered, pos... | 3ojHTwrvx90 | The video depicts President Park Geun-hye's response to a series of strong earthquakes in the Gyeongju region of Gyeongbuk. The President instructs that the area be considered for declaration as a special disaster area and orders a thorough review of the disaster safety system, including earthquake prevention measures. She visits the Wolsong Nuclear Power Plant, emphasizing the importance of rigorous earthquake countermeasures in nuclear power. The President is seen walking with officials, all dressed in yellow jackets, through public areas, assessing damage and discussing recovery measures. The officials appear engaged and attentive, highlighting the seriousness of the situation. The video also shows news broadcasts discussing the President's instructions and the consideration of declaring Gyeongju a special disaster area, which would provide relief to affected residents by alleviating the burden of damage recovery and offering benefits such as reduced public utility fees. Throughout the video, the focus remains on the President's efforts to ensure comprehensive safety measures are in place and to provide support to the affected region. | {
"original": " 박근혜 대통령이 최근 강진이 잇따라 발생한 경북 경주지역을 찾아 특별재난지역 선포를 적극 검토하라고 지시했습니다. 이어 지진 등 재난안전시스템을 원점에서 재점검할 것도 주문했습니다. 박 대통령은 아울러 인근 월성원자력발전소를 찾아 원전은 단 한 치의 실수도 용납되지 않는다며 지진 방지 대책을 철저히 재점검해달라고 당부했습니다. 박 대통령의 지시에 국민안전처 또한 경주시를 특별재난지역으로 선포할 예정입니다. 특별 재난지역으로 지정되면 지자체 피해 복구비 부담이 완화되고 피해 주민들은 공공요금 감면 혜택을 받게 됩니다.",
"refined": "President Park Geun-hye has ordered the declaration of a special disaster area to be actively reviewed in the Gyeongbuk-Gyeongju region, which has recently been hit by a series of earthquakes. She also ordered a thorough review of the disaster safety system, including earthquake preparedness. President Park visited the nearby Wolseong Nuclear Power Plant, stressing that not even a single mistake can be tolerated in nuclear power plants, and called for a thorough re-examination of earthquake prevention measures. The National Safety Agency is also planning to declare Gyeongju a special disaster area. If designated as such, local governments will see a reduced burden in damage recovery costs, and affected residents will receive benefits such as reduced public utility fees.",
"summarized": "President Park Geun-hye has ordered a review of declaring the Gyeongbuk-Gyeongju region, which was recently hit by a series of earthquakes, as a special disaster area. She also instructed for a thorough re-examination of the country's disaster safety system, including earthquake measures. Additionally, she visited a nearby nuclear power plant and emphasized the importance of a thorough review of earthquake prevention measures, stating that not even one mistake is tolerable. If declared a special disaster area, the local government's burden of damage recovery will be alleviated, and victims will be eligible for benefits such as reduced public fees.",
"translated": "President Park Geun-hye has instructed that the declaration of a special disaster area be actively considered for the Gyeongju region in Gyeongbuk, which has recently experienced a series of strong earthquakes. She also ordered a thorough review of the disaster safety system, including earthquake prevention measures, from the ground up. Additionally, President Park visited the nearby Wolsong Nuclear Power Plant and emphasized that not even a single mistake can be tolerated in nuclear power, requesting a thorough re-examination of earthquake countermeasures. Following the President's instructions, the National Safety Agency plans to designate Gyeongju as a special disaster area, which would alleviate the burden of damage recovery for local governments and provide benefits such as reduced public utility fees for affected residents.",
"translated_llm": "President Park Geun-hye has ordered that the special disaster zone be actively reviewed in the area of the recent strong earthquake. He also ordered that the earthquake and other disaster prevention systems be reviewed at the source. The President also requested that the nearby Wulan-Sun power plant be reviewed and that the plant not even one mistake be allowed.",
"translated_whisper": " President Park Geun-hye has ordered the special disaster area declaration to be actively reviewed in the Gyeongbuk-Gyeongju region, which has recently been hit by a strong earthquake. He also ordered the disaster safety system such as earthquakes to be re-examined at the source. President Park said that he would look for a nearby monthly nuclear power plant and ask for a thorough re-examination of the earthquake prevention measures, saying that the nuclear power plant is not allowed to make even one mistake. The National Safety Agency is also planning to declare Gyeongju as a special disaster area. If designated as a special disaster area, the burden of damage recovery will be relieved and the victims will receive public funds."
} |
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... | [
"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 rubb... | [
"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 ea... | {
"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"
} | 32 | [
"The frame shows a news presenter dressed in traditional attire standing in front of a cityscape backdrop. The presenter is delivering news about a recent earthquake in Gyeongju, which resulted in 23 injuries and significant property damage. The text at the bottom of the screen provides details about the earthquake... | w6jdwGYcKZU | A news report covers a 5.8 magnitude earthquake in Gyeongju, resulting in 23 injuries and 1,118 cases of property damage. The majority of the injured are from Gyeongbuk, with others from Ulsan, Daegu, and other regions. Seven of the injured have been released from the hospital, while 16 remain hospitalized. The safety agency plans to conduct safety inspections and assess the structural integrity of damaged homes to determine if it is safe for residents to return. The report shows footage of damaged buildings, debris-filled streets, and rescue efforts, highlighting the extent of the destruction and the need for ongoing safety measures. The community is responding to the disaster, with emergency services and inspections underway to ensure the safety of residents in the affected areas. | {
"original": " 국민안전처는 경주에서 발생한 규모 5.8의 지진으로 부상자 23명이 발생했고 재산상 피해는 1118건으로 집계됐다고 밝혔습니다. 안전처와 보건복지부에 따르면 부상자 23명 가운데 7명은 병원에서 치료를 받고 귀가했으며 16명은 입원 중입니다. 지역별 부상자는 경북이 12명으로 가장 많았고 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.",
"refined": "The National Security Agency reported that 23 people were injured in the 5.8 magnitude earthquake in Gyeongju, with 1,118 cases of property damage. According to the Ministry of Health and Welfare, of the 23 injured, 7 were treated at the hospital and have been discharged, while 16 remain hospitalized. By region, Gyeongbuk had the most injured with 12, followed by 4 in Ulsan, 2 in Daegu, and 1 in Busan, Incheon, Chungbuk, Jeonnam, and Gyeongnam each. The safety office plans to conduct safety inspections with rescue technicians for homes with cracks and review their habitability and additional safety measures.",
"summarized": "A 5.8 magnitude earthquake in Gyeongju resulted in 23 injuries, with 16 people hospitalized and 7 treated and released. The majority of the injured were from Gyeongbuk, with 12 cases, followed by 4 in Ulsan, 2 in Daegu, and 1 each in Busan, Incheon, Chungbuk, Jeonnam, and Gyeongnam. Damage was reported in 1,118 cases. The National Security Agency will conduct safety inspections and review additional safety measures for homes with cracks.",
"translated": "The National Security Agency reported that 23 people were injured in the 5.8 magnitude earthquake in Gyeongju, with property damage estimated at 1,118 cases. According to the Ministry of Health and Welfare, 7 out of the 23 injured individuals were treated at the hospital and have since returned home, while 16 remain hospitalized. The majority of the injured, 12 people, were from Gyeongbuk, followed by 4 in Ulsan, 2 in Daegu, and 1 each in Busan, Incheon, Chungbuk, Jeonnam, and Gyeongnam. The safety agency plans to conduct safety inspections, along with structural engineers, for homes with damage, and will assess whether it is safe for residents to return and implement additional safety measures as needed.",
"translated_llm": "The National Security Agency said that 23 people were injured in the 5.8 magnitude earthquake in 경주 and that property damage was estimated at 1,118. According to the Ministry of Safety and Health, seven of the 23 injured are being treated in hospital and 16 are in hospital. The number of injured in the area is the highest at 12 people, with four in Ulsan, two in Daegu, Busan, Incheon, 충북, 전남, and one in 경남.",
"translated_whisper": " The National Security Agency reported that 23 people were injured due to the 5.8 magnitude earthquake in Gyeongju, and that the damage was 1118. According to the Ministry of Health and Welfare, 7 out of 23 people were treated at the hospital and 16 were hospitalized. The number of injured people in each region was the largest with 12 people in Gyeongbuk, 4 in Ulsan, 2 in Daegu, and 1 in Busan, Incheon, Chungbuk, Jeonnam, and Gyeongnam. The safety office will check safety with the rescue technicians and other additional safety measures for homes with a"
} |
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... | [
"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 ... | [
"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 responder... | {
"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"
} | 32 | [
"In the current frame, a person is seen inside a convenience store, bending over while inspecting or looking at items on the lower shelves. The setting appears to be during a housewarming preparation, as suggested by the ASR transcript. The individual seems focused on selecting or checking specific items among the ... | oK2O8Xtkrvs | A person is preparing for a housewarming party, initially searching for items in a convenience store. The scene then shifts to the aftermath of an earthquake, where the convenience store is in disarray with items scattered across the floor and shelves partially emptied. The video continues to show the chaotic scene, emphasizing the impact of the earthquake, before concluding with an end screen displaying the "OhmyTV" logo and credits. The original ASR transcript mentions a housewarming party preparation and a conversation, but it does not directly relate to the earthquake scene, suggesting the video may be a compilation or have a narrative that combines different events. | {
"original": " 오늘 저녁은 아침식사! 집들이 집들기 중이라 집들기는 간단해요 우리 집 주변에 있는 집들기 집들이 집들기 준비해놓은 자석을 위해 저의 집을 준비하려는 집들이 집으로 돌아간 뒤 아, 엄마 이거... 엄마, 여보 여기 있는 집! 아, 엄마는 어떻게 그래?? 아가... 아, 마마도가 없었다고...",
"refined": "This evening's dinner is like breakfast. Since we're in the middle of moving, the moving is simple. I'm preparing my house for the housewarming party that the neighboring houses are hosting, so I went back to my house and said, \"Mom, look at this house! Ah, why are you like this, Mom?\" My mom wasn't there...",
"summarized": "Not Available",
"translated": "Tonight's dinner is breakfast. Since we're in the middle of moving, the housewarming party preparations are simple. I went back to my house to prepare for the housewarming party, which is being held at a nearby house that we've prepared for. Oh, Mom, look at this... Mom, look, there's a house here. Oh, why is Mom like that? The baby... Oh, Mama isn't here...",
"translated_llm": "This evening is breakfast! The houses are being cleaned and cleaned are simple. The houses around our house are cleaned and cleaned. The houses around our house are cleaned and cleaned.",
"translated_whisper": " The shooting scene was taken by B.A.T. I couldn't see B.A.T more carefully However, the camera was already on My camera's still on B.A.T was happy with the shooting"
} |
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... | [
"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 ... | [
"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 responder... | {
"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"
} | 32 | [
"The current frame shows a white pickup truck parked on a residential street, surrounded by greenery and autumn foliage. The scene appears to be set in a quiet neighborhood with a paved road leading through the property. The text on the image indicates an event of interest related to the truck, mentioning a locatio... | 6HhGFKbHFbs | A kidnapping incident involving a 5.4-year-old child occurred in Gyeongbuk on November 15th. The police have started investigating the incident and are working to locate the child. The community is involved in the search, with people walking around and looking for the child. The scene is set in a quiet neighborhood with greenery and autumn leaves, and later shifts to a busy street with various storefronts, including a "Super Size" store and a clothing shop. Pedestrians are walking along the sidewalk, and some are using their phones. The police investigation and community efforts are ongoing, with the scene showing people moving about, some in a hurry, and others standing near storefronts, looking at their phones. The atmosphere remains active and urban, with various shops lining the street and people engaged in their daily activities. The video concludes with a production information screen from Hankook Ilbo, indicating the end of the segment. | {
"original": " Thank you.",
"refined": "Thank you.",
"summarized": "Not Available",
"translated": "Thank you.",
"translated_llm": "- Thank you. - Thank you.",
"translated_whisper": " The The The The The The The The The The The The The The Thank you for watching!"
} |
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... | [
"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 ... | [
"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 responder... | {
"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"
} | 32 | [
"The image shows the exterior of a modern-looking building with glass doors and windows, located in Kyongbok, Pohang, approximately 7 kilometers north of a station. The text on the image indicates that an earthquake with a magnitude of 5.4 occurred in the central region, causing ground shaking, damage to structures... | 1sXMvTYhnn0 | The video depicts the aftermath of an earthquake with a magnitude of 5.4 that occurred in Pohang, North Gyeongsang, approximately 7 kilometers north of a station. The earthquake caused ground shaking, damage to structures, and potential loss of electricity. The video shows exterior and interior scenes of damaged buildings, including broken glass and fallen objects such as wall clocks and water containers. The footage also captures street scenes, likely taken from a car, showing pedestrians and storefronts with Korean signage, indicating the disruption to daily life. Additionally, the video covers the impact on transportation, mentioning that four trains were overturned in Daegu, and shows scenes from a train station or transportation hub, highlighting efforts to ensure passenger safety. The video concludes with a series of frames showing the interior of a structure with a glass ceiling, possibly assessing the damage, followed by end credits acknowledging the contributors and source of the video footage. The overall context suggests a news report covering the earthquake's effects on the region, including damage, disruption, and response efforts. | {
"original": "JR東日本E233系電車大阪駅 新大阪線大阪駅の駅は、大阪駅の中央駅にある駅です。新幹線E2系電車",
"refined": "JR East Japan's E233 series train at Osaka Station's Shin-Osaka Station is a central station in Osaka.",
"summarized": "Not Available",
"translated": "The JR East E233 series train at Osaka Station, the Osaka Station on the new Osaka line is a station located in the central station of Osaka Station, near the Shinkansen E2 series train.",
"translated_llm": "JR東日本E233系電車大阪駅 新大阪線大阪駅の駅は、大阪駅の中央駅にある駅です。新幹線E2系電車",
"translated_whisper": " I came to a convenience store in the morning. I have a lot of things to do, so I'm going to take a video. I'm going to eat a lot of delicious food. I'm going to take a video of this. It's a movie called Tietou. I will take a walk on this street. The T-Mobile in Shibuya station has been sold off."
} |
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... | [
"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 ... | [
"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 responder... | {
"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"
} | 32 | [
"The image is a map highlighting a specific area with a red target symbol. The surrounding regions are labeled in Korean: \"포항시북구\" (Pohang-si Buk-gu), \"포항시남구\" (Pohang-si Nam-gu), \"영천시\" (Yeongcheon-si), and \"경주시\" (Gyeongju-si). The map suggests a focus on Pohang-si Buk-gu, indicated by the target symbol, poss... | MH6_TeWDitQ | The video appears to be a news broadcast from Yonhap News TV, focusing on a specific event or incident in Pohang-si Buk-gu, South Korea. The majority of the frames display a map highlighting this region, with a red target symbol, suggesting the location of the event. The surrounding areas, including Pohang-si Nam-gu, Yeongcheon-si, and Gyeongju-si, are also visible on the map.
The latter part of the video shifts to footage of a street, initially showing a group of people gathered at a crosswalk, waiting to cross. The atmosphere is orderly, and the crowd size increases over several frames, indicating a growing presence. The "Yonhap News TV" text is consistently visible, confirming the context of the footage as a news broadcast.
The scene then changes to depict the aftermath of an incident, with debris scattered across the road. The atmosphere is calm, with minimal activity or movement. An individual is seen standing on the sidewalk, observing the situation. The urban setting, characterized by multi-story buildings and overhead power lines, remains consistent throughout these frames.
Despite the "Thank you for watching!" at the beginning of the ASR transcript, which seems out of place, the video's content suggests it is covering an event or incident in Pohang-si Buk-gu, showing both the geographical context through maps and the event itself through street footage, ending with the depiction of an incident's aftermath. | {
"original": " Thank you for watching!",
"refined": "Thank you for watching.",
"summarized": "The speaker is concluding a segment or show and is either thanking the audience or indicating a brief pause or break.",
"translated": "Thank you for watching!",
"translated_llm": "Thank you for watching!",
"translated_whisper": " We'll be right back."
} |
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... | [
"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 ... | [
"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 responder... | {
"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"
} | 32 | [
"The current frame shows an aerial view of a city with a network of roads and highways, tall buildings, and a mix of residential and commercial areas. The visible text \"15일 오후 2시 29분\" indicates the time is the 15th day at 2:29 PM. The scene likely serves as a backdrop for the person taking photos, as mentioned in... | _HN3DpC6m6I | The video begins with aerial views of a city, specifically Pohang, North Gyeongsang, where a 5.4 earthquake has occurred. The initial frames show the urban landscape, emphasizing the city's infrastructure and possibly the aftermath of the earthquake. The scene then transitions to a person taking photos from a secluded spot, as mentioned in the ASR transcript.
As the video progresses, it shifts to a parking area under a canopy, showing cars and a calm environment, which contrasts with the earlier depiction of the earthquake's impact. The footage then moves indoors, focusing on everyday objects like a large water bottle and a bowl with liquid, possibly highlighting the domestic environment and daily routines.
However, the scene abruptly changes to a store with a broken glass window or door, indicating a disruption or possibly looting. The store appears to be a commercial establishment, with signs advertising products like the iPhone X. This disruption contrasts with the previous calm scenes.
The video then cuts to outdoor public areas, such as parks or plazas, where large groups of people are gathered. These gatherings seem to be in response to warnings from the meteorological agency about aftershocks and safety precautions. The atmosphere is one of concern and attentiveness, with people discussing and reacting to the advisory.
The final parts of the video continue to show these gatherings, with people interacting and possibly receiving information about safety measures. The setting remains organized, and the participants appear focused on the information being shared. Throughout the video, the context shifts from the urban landscape and personal, secluded moments to public gatherings focused on safety and response to natural disasters. | {
"original": " In this video, I will show up in a cool place where no one can see by myself. It's a day like this. In fact, he will take a photo. I thought he will be able to take photos at home. Today, I'm in the bus. Takk for at du så.",
"refined": "I will be in a cool place by myself where no one can see me. It's a day like this. In fact, I will take a photo. I thought I would be able to take photos at home. Today, I'm on the bus. Thank you.",
"summarized": "Not Available",
"translated": "In this video, I will show up in a cool place where no one can see by myself. It's a day like this. In fact, he will take a photo. I thought he would be able to take photos at home. Today, I'm in the bus. Thanks for watching.",
"translated_llm": "In this video, I will show up in a cool place where no one can see by myself. It's a day like this. In fact, he will take a photo. I thought he will be able to take photos at home. Today, I'm in the bus.",
"translated_whisper": " I feel so happy to watch a shooting star wearing yellow clothes and a beautiful smile in a photo. I enjoyed filming in the summer and now I think I'm a bit sad. I was told how to pose in the photos. By the way, the reason I picked this shoot is because... The second most famous part of the concert is a pool."
} |
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... | [
"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 ... | [
"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 responder... | {
"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"
} | 32 | [
"The current frame shows a news anchor, likely delivering a news segment. The focus is on an event organized by the Korea Forest Service, where they are providing free forest fuel to Pohang residents affected by the earthquake. The support is aimed at earthquake victims, lone elderly individuals, and those receivin... | bADRfbUZewY | The video depicts an event organized by the Korea Forest Service to provide free forest fuel to Pohang residents affected by an earthquake. The support targets earthquake victims, lone elderly individuals, and basic livelihood security beneficiaries, with each of the 21 households receiving one ton of firewood. The firewood is sourced from oak, birch, and Quercus serrata trees produced in the Geumgang Sunae Forestry Management District in Bonghwa County's Janggunbong model forest management area. The event is held at the Pohang Forest Cooperative Wood Distribution Center, where individuals in red jackets are seen loading, unloading, and organizing the firewood for distribution. The video shows a news anchor discussing the initiative, as well as the exterior of the Forest Service building and the distribution center, highlighting the collaborative efforts to efficiently provide support to the affected households. Overall, the video showcases the Korea Forest Service's efforts to provide aid to those in need following the earthquake. | {
"original": " 산림청은 오는 4일 포항산림조합목재유통센터에서 구규림에서 생산된 임산연료를 지진피해를 입은 포항주민들에게 무상으로 공급하는 행사를 갖기로 했습니다. 지진피해 주민과 홀몸노인, 기초생활보장수급자 등을 대상으로 가구당 1톤씩 모두 21가구를 지원합니다. 이번에 공급하는 장작은 봉화군 장군봉 선도산림경영단지 내 금강소나무육성사업지에서 생산된 참나무와 벗나무, 박달나무 등입니다.",
"refined": "The Korea Forest Service will hold an event at the Pohang Forestry Cooperative Lumber Distribution Center on the 4th to provide free forest fuel produced in Gugyurim to Pohang residents who were affected by the earthquake. The support targets include earthquake victims, elderly living alone, and basic living security recipients, with one ton of fuel per household for a total of 21 households. The firewood to be supplied this time is oak, birch, and hackberry trees produced in the Geumgang Sonaemun Forestry Business Area in Bonghwa County's Janggunbong Forest Management District.",
"summarized": "The Salimcheon will supply 1 ton of firewood to 21 households, including earthquake victims, elderly individuals, and basic living insurance recipients in Pohang, free of charge. The firewood, consisting of various types of trees, was produced in the Geumgang Sonaemun Forestry Business Area in Bonghwa County.",
"translated": "The Korea Forest Service will hold an event at the Pohang Forest Cooperative Wood Distribution Center on the 4th to provide free forest fuel produced in Gugyurim to Pohang residents affected by the earthquake. The support targets include earthquake victims, lone elderly individuals, and basic livelihood security beneficiaries, with one ton of fuel provided per household, totaling 21 households. The firewood to be supplied this time is made from oak, birch, and Quercus serrata trees produced in the Geumgang Sunae Forestry Management District in Bonghwa County's Janggunbong model forest management area.",
"translated_llm": "The Sanforum has decided to hold an event to provide free of charge the maternity fuel produced in the 구규림 by the sanforum in the next 4 days. The sanforum supports all 21 households for the purpose of providing 1 ton per household for the residents of the geyn피해 and holbod노인, 기초생활보장수급자 etc. The main products supplied this time are the sanforum and palm trees produced by the sanforum in the national gold mining company.",
"translated_whisper": " The Salimcheon will have a event to supply the labor fuel produced in Gugyurim to the poor residents who were affected by the earthquake on the 4th of this year. We will provide 21 furniture per 1 ton for the victims, the elderly, and the basic living insurance providers. The materials supplied are pine trees, pine trees, and pine trees that are produced in Geumgang Sonaemun Forestry Business Area."
} |
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... | [
"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 ... | [
"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 responder... | {
"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"
} | 32 | [
"The image shows a scene cordoned off with yellow police tape, indicating restricted access. There is visible damage to a structure in the background, suggesting the aftermath of an incident, possibly a fire or explosion. The text in the center of the image reads \"죄송합니다\" (We apologize), which might be related to ... | 4eNANKub4NE | The video depicts the aftermath of a devastating incident, likely an earthquake, which has caused significant destruction to buildings and infrastructure. The scene is set in a rural or semi-rural area, with damaged structures, debris, and rubble scattered everywhere. The video shows individuals reflecting on the disaster, seeking guidance, and praying for salvation and divine grace. A church plays a central role in the community's response, serving as a place for prayer, reflection, and communal support. The video features testimonials from individuals who have been affected by the disaster, sharing their emotional experiences and expressing gratitude for God's grace. Throughout the video, there is a strong emphasis on community, faith, and the importance of spiritual connection in the face of adversity. The church is portrayed as a symbol of hope and a source of comfort, where people can come together to pray, seek support, and find strength in their shared beliefs. The video concludes with images of the church and its tower, featuring a cross and a Star of David, highlighting the church's role in providing spiritual guidance and support to its diverse community. | {
"original": " PANDA PANDA PANDA PANDA PANDA PANDA The I cried the most when I met my friend. I prayed for her to be saved. I pray for you. God, you made us live happily. It's God's grace. EJ, now.",
"refined": "I cried the most when I met my friend. I prayed for her to be saved. I pray for you. God, you made us live happily. It's all thanks to God.",
"summarized": "The speaker expresses gratitude for a happy life, attributing it to God's grace. They recall a emotional experience where they cried the most, likely when a friend or someone they care about was rescued or safe, and emphasize the importance of prayer.",
"translated": "I cried the most when I met my friend. I prayed for her to be saved. I pray for you. God, you made us live happily. It's God's grace. EJ, now.",
"translated_llm": "I cried the most when I met my friend. I prayed for her to be saved. I pray for you. God, you made us live happily. It's God's grace. EJ, now.",
"translated_whisper": " Hello everyone! I'm going to make a very special cake today. I'm going to use a cake sheet that I made before. I'm going to use a cake sheet that I made before. I'm going to use a cake sheet that I made before. The mother gets upset But when she gets there It means that they are safe She's finally rescued I got the most tears, it says I can only cry Lee Jang-ro, you should pray and serve him We pray together God, you made us live happily It's all thanks to God"
} |
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... | [
"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 ... | [
"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 responder... | {
"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"
} | 32 | [
"In the current frame, a group of people is running away from a cloud of smoke on a street, suggesting a sudden emergency or disturbance. The individuals appear to be in a hurry, with some looking back towards the source of the smoke. The setting includes a tree-lined street and buildings in the background, indicat... | QIkpgLcEWMA | The video appears to depict a series of scenes related to an emergency situation, likely an earthquake, in an urban and rural setting. The initial frames show people running away from a cloud of smoke, indicating a sudden disturbance. As the video progresses, it transitions to show damaged buildings, rubble, and debris, with text overlays suggesting ongoing aftershocks and the need for caution. The setting shifts between urban streets, a salon, and rural areas with damaged buildings. The video also includes footage of people eating and conversing in a casual setting, as well as CCTV views of parking areas with no visible activity. Throughout the video, text overlays provide context, including "소셜 스토리" (Social Story) and "시청자제보" (Viewer Report), indicating that the footage is part of a news segment or social media report. The video concludes with a title screen and a transition to a JTBC logo, suggesting a broadcast or network introduction. The original ASR transcript mentions an article and the need for something "to go," but its relevance to the video content is unclear. Overall, the video seems to document the aftermath of an earthquake or similar disaster, with a mix of emergency responses, damage assessments, and casual conversations. | {
"original": " 또 다른 기사 또 다른 기사 뭐야 뭐가 이거 야XXX GZ는 그건데 이거 나가야 돼, 나가야 돼 나가야 돼, 나가야 돼 XXX야 다음 영상에서 만나요.",
"refined": "It looks like another article, but what is this? Anyway, GZ is ready to go out. See you in the next video.",
"summarized": "The speaker appears to be wrapping up a video or live stream, mentioning that the team has completed a mission and thanking the viewers for watching, with an invitation to meet again in the next video.",
"translated": "Another article, another article, what's this? XXX GZ, that's it, but this has to go, has to go, has to go, has to go, XXX, see you in the next video.",
"translated_llm": "Another article another article what is this or XXX GZ is that but this is going to go, going to go, going to go, going to go XXX is See you in the next video.",
"translated_whisper": " 2X2 3X2 In a very fast forward The whole team has already started to attack What is that The team has already prepared a long time, but... The mission to the house is now complete Thank you. Thank you for watching!"
} |
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... | [
"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 ... | [
"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 responder... | {
"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"
} | 32 | [
"The current frame shows a news broadcast with two news anchors, a man and a woman, seated at a desk. They appear to be delivering a report on a recent earthquake. The backdrop displays a cityscape under a clear sky, suggesting an urban setting. Text at the bottom of the screen indicates the earthquake's magnitude ... | _vMNNh-x79Q | A news broadcast reports on a 5.4 to 5.5-magnitude earthquake that occurred in Pohang, South Korea, specifically 6 kilometers north of Buk-gu. The quake caused significant shaking, with reports of damage, aftershocks, and evacuations in various parts of the city and beyond, including Seoul and Busan. The Korea Meteorological Administration warned of potential aftershocks and urged caution. Despite the earthquake, the Ministry of Education announced that the CSAT exam would proceed as scheduled nationwide. The nuclear power plant operator, Hanwha, reported that all plants were operating normally without issues. The earthquake was felt nationwide, prompting national concern, and the exact number of casualties and extent of damage were not immediately confirmed. The news report provided updates on the situation, including the impact on daily life, infrastructure, and the government's response to the disaster. | {
"original": " 지금 저희가 이와 관련해서 보도국 취재기자를 연결해서 자세한 내용을 좀 들어보도록 하겠습니다. 홍정원 기자. 네, 오늘 오후 2시 29분 포항 북구 북쪽 6km 지역에서 규모 5.5의 지진이 발생했다고 기상청이 밝혔습니다. 진앙은 북위 36.10도, 동경 129.37도입니다. 포항에서는 창이 흔들리고 책상에서 책이 떨어지는 등 진동의 여파가 상당했던 것으로 전해지고 있는데요. 또한 포항 곳곳에선 여진도 이어진 것으로 전해지고 있습니다. 기상청은 이동속도가 빠른 지진파, 피파만을 이용해 자동추정한 정보라며 여진 등 안전에 주의해야 한다고 당부했습니다. 한수원 측은 전국의 원자력발전소는 별다른 문제없이 가동중이라고 밝혔습니다. 오늘 지진의 지진은 서울 등 경기 남부는 물론 경남권과 호남권 등 전국적으로 감지된 것으로 전해지고 있습니다. 지진 발생 직후 부산소방안전본부와 부산경찰청에는 건물이 흔들린다는 시민들의 문의전화가 빗발친 것으로 전해졌습니다. 일부 도심 건물이 심하게 흔들리면서 관련 직장인과 시민들이 긴급 대피하는 소동이 벌어졌습니다. 기상청은 북구로부터 북쪽 9km 지역에서도 규모 5.4 지진이 추가 발생했다고 밝혔습니다. 다시 한번 말씀드리겠습니다. 오늘 오후 2시 29분 포항 북구에서 북쪽으로 6km 지역에서 규모 5.5의 지진이 발생했다고 기상청이 밝혔습니다. 포항 곳곳에선 여진이 이어지면서 주민대피가 속출했으며 서울 등 경기권과 부산에서도 지진이 감지되면서 전국적으로 우려를 키우고 있습니다. 원자력발전소를 관리하는 한수원 측은 현재까지 가동 중인 원전에 별다른 문제가 없다고 밝혔습니다. 아직까지 전국적으로 정확한 인명, 시설 피해 상황은 집계되지 않은 상황입니다. 지금까지 보도국에서 전해드렸습니다.",
"refined": "Let's connect with our newsroom reporter, Hong Jungwon, for more information. Yes, the weather agency reported that a 5.5 magnitude earthquake occurred at 2:29 pm today, 6 km north of Pohang's North District. The epicenter is located at 36.10 degrees north latitude and 129.37 degrees east longitude. In Pohang, the quake caused significant shaking, with reports of windows rattling and books falling off desks. There have also been reports of aftershocks in various parts of the city. The weather agency cautioned that, based on the fast-moving seismic wave, there may be further aftershocks, and people should exercise caution. Hanwha, the operator of the country's nuclear power plants, stated that all plants are operating normally. Today's earthquake was felt nationwide, including in Seoul, the southern part of Gyeonggi Province, and the Gyeongsangnam and Honam regions. Following the quake, the Busan Fire Department and Busan Police Agency received numerous calls from citizens reporting shaking buildings. Some buildings in the city center shook violently, prompting office workers and residents to evacuate in a state of panic. The weather agency reported another earthquake, with a magnitude of 5.4, occurring 9 km north of Pohang's North District. To recap, a 5.5 magnitude earthquake struck 6 km north of Pohang's North District at 2:29 pm today, causing widespread shaking and prompting evacuations in various parts of the city. The quake has raised concerns nationwide, with reports of shaking in Seoul, Gyeonggi Province, and Busan. Hanwha stated that there are currently no issues with the operating nuclear power plants. As of now, there is no confirmed information on casualties or damage to facilities.",
"summarized": "A 5.5 magnitude earthquake occurred in Pohang, North Gyeongsang Province, at 2:29 pm, with its epicenter 6km north of the city. The earthquake was felt nationwide, including Seoul, Gyeonggi Province, and Busan, causing concerns. The National Weather Agency warned of potential aftershocks and advised caution. Hanwha, the operator of the country's nuclear power plants, reported that all plants were operating normally. Residents in Pohang were evacuated, and some buildings in the city were severely shaken, but no casualties or facility damage have been reported so far.",
"translated": "We will now connect with our reporter, Hong Jungwon, from the newsroom to provide more detailed information. Yes, at 2:29 pm today, the Korea Meteorological Administration announced that a 5.5-magnitude earthquake occurred in an area 6 kilometers north of Pohang's Buk-gu. The epicenter was located at 36.10 degrees north latitude and 129.37 degrees east longitude. In Pohang, the quake caused significant shaking, with reports of windows rattling and books falling off desks. Additionally, there have been reports of aftershocks in various parts of the city. The Korea Meteorological Administration warned that the quake's fast-moving seismic waves were automatically detected, and urged caution regarding safety and potential aftershocks. Hanwha, the operator of the country's nuclear power plants, reported that all plants are operating normally without any issues. The earthquake was felt nationwide, including in Seoul, the southern part of Gyeonggi Province, and the Gyeongsang and Honam regions. Immediately after the quake, the Busan Fire Department and Busan Police Department received numerous calls from citizens reporting shaking buildings, and some office workers and citizens in the city center were forced to evacuate due to violently swaying buildings. The Korea Meteorological Administration also reported a 5.4-magnitude quake in an area 9 kilometers north of Buk-gu. To repeat, at 2:29 pm today, a 5.5-magnitude earthquake occurred in an area 6 kilometers north of Pohang's Buk-gu, according to the Korea Meteorological Administration. Aftershocks have been reported in various parts of Pohang, prompting evacuations, and the quake has been felt in Seoul, Gyeonggi Province, and Busan, sparking national concern. Hanwha, the nuclear power plant operator, has reported that all plants are operating normally. As of now, the exact number of casualties and extent of damage nationwide have not been confirmed. This is the end of our report from the newsroom.",
"translated_llm": "The weather service reported that a magnitude 5.5 earthquake occurred in a region 6km north of Port-au-Prince at 2h29m today. The quake was reported to have caused considerable damage in Port-au-Prince, including windows shaking and books falling from the desk. There were also reports of earthquakes throughout the city. The weather service said that the speed of the quake was very fast, using automatic information, the quake was reported to be safe.",
"translated_whisper": " Let's listen to the news from the newsroom. Reporter Hong Jungwon. Yes, today at 2.29, the weather forecast revealed that a 5.5 magnitude earthquake occurred in the 6km area of Pohang, north of Pohang. The current temperature is 36.10 degrees north, and the temperature of the east is 129.37 degrees. It is said that the vibration was quite severe, such as shaking windows and books falling off the desks. The news also said that the earthquake has been reported to be repeated in all areas. The National Weather Agency said that the information was automatically corrected by using only the P-train, which has a fast movement speed, and that safety such as the earthquake should be taken care of. Han Su-won said that the nuclear power plant in the whole country is running without any problems. The earthquake of today's earthquake is reported to be nationwide, including Seoul and the southern part of Gyeonggi, as well as Gyeongsangnam-do and Honang-do. The The weather service said that an earthquake occurred 5.5 times in the 6km area from Pohang North District at 2.29 pm today. The earthquake was still in place in various parts of Pohang, and the residents were evacuated. Earthquakes were also detected in Gyeonggi-do and Busan, including Seoul, and are causing national concern. Han Suwon, who manages the nuclear power plant, said that there is no problem with the nuclear power plant operating so far. The situation of the exact name and the damage to the facility has not been discovered yet. I've been telling you from the news agency."
} |
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... | [
"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 ... | [
"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 responder... | {
"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"
} | 32 | [
"In the current frame, a news anchor is delivering a report about the aftermath of an earthquake in Pohang. The anchor is dressed formally in a suit and tie, indicating a professional broadcast setting. Behind him, there is a split-screen showing a scene of destruction, including damaged apartment buildings with vi... | k_iUdIJKUUc | The video depicts the aftermath of a 5.4-magnitude earthquake in Pohang, South Korea, where over 700 citizens are unable to return home due to significant damage to apartment buildings. The news anchor delivers a report on the situation, showing footage of damaged buildings with cracks in the walls, shattered windows, and displaced residents taking shelter at the Heunghae Indoor Sports Center, a temporary evacuation center set up by the city of Pohang. Reporter Ha Dong-won provides live updates from the scene, highlighting the fear and concerns among residents about potential building collapses. The video shows residents gathered at the sports center, receiving emergency relief supplies, including water, ramen, and daily necessities. The report emphasizes the scale of the displacement, with over 730 people affected, and the city's response in providing temporary shelter and support. The footage also captures the urgency and chaos of the situation, with people evacuating their homes, carrying belongings, and seeking refuge. The video concludes with a summary of the situation, highlighting the ongoing concerns about building safety and the need for continued support for the affected community. | {
"original": " 포항 시민들은 눈앞에서 아파트 외벽이 갈라지는 현장을 목격하면서 하루 종일 공포에 떨었습니다. 시민 700여 명은 지금도 집에 돌아가지 못하고 대피소에 머물고 있습니다. 현장 연결합니다. 하동원 기자, 거기가 어느 대피소입니까? 네, 포항시가 임시 주민대피소로 마련한 흥해실내체육관입니다. 이곳에는 지난주였던 포항 북구 주민을 비롯한 500여 명이 피신해 있는 상태입니다. 규모 5.4 지진 발생 후 여진이 계속 이어지면서 주민들의 불안감은 가시지 않고 있습니다 일부 지역에선 아파트 건물 외벽이 갈라지면서 내부 철근까지 밖으로 노출됐고 출입문 유리창이 산산조각 나는 등 건물 붕괴 가능성까지 제기됐기 때문입니다 포항시는 재난대책본부는 지금까지 730여 명의 이재민이 발생했다고 밝혔습니다 대다수 주민들이 생필품조차 챙기지 못하고 피신했습니다 기업과 단체는 생수, 라면, 생활용품 등 긴급 구호 물품을 지원하고 있습니다 국민안전처는 지난해 12월 전국에 지진대피소 7,068곳을 지정했습니다 지진대피소 위치와 현황은 국민재난안전포탈과 안전디딤돌 스마트폰 애플리케이션 등에서 확인할 수 있습니다 지금까지 경북 포항에서 TV조선 하동훈입니다",
"refined": "The citizens of Pohang witnessed the exterior walls of apartment buildings cracking right before their eyes and were terrified throughout the day. Over 700 residents are still unable to return home and are staying in shelters. Let's go live to the scene. Reporter Ha Dong-won, which shelter is that? Yes, it's the Henghae Indoor Sports Center, which the city of Pohang has set up as a temporary shelter. Currently, over 500 people, including residents of Pohang's North District, are taking refuge here. Following the 5.4-magnitude earthquake, aftershocks have continued, and residents' anxiety has yet to subside. In some areas, the exterior walls of apartment buildings have cracked, exposing internal steel frames, and entrance doors and windows have shattered, raising concerns about building collapse. The Pohang disaster response headquarters has reported that over 730 people have been displaced. The majority of residents fled without even packing essentials. Companies and organizations are providing emergency relief supplies, including water, instant noodles, and daily necessities. Last December, the National Emergency Management Agency designated 7,068 earthquake shelters nationwide. The locations and status of these shelters can be found on the National Disaster Safety Portal and the Safety Dream DOL smartphone application. This is Ha Dong-won reporting from Pohang, Gyeongbuk, for TV Chosun.",
"summarized": "A 5.4 magnitude earthquake struck Pohang, causing widespread fear among citizens who witnessed apartment buildings cracking and partial collapses. Approximately 730 people, including residents of Pohang's North District, have taken shelter, with many unable to return home due to concerns about building safety. The city has set up temporary shelters, including the Henghae Indoor Sports Center, which is currently housing around 500 people. Relief efforts are underway, with companies and organizations providing essential supplies such as water, food, and daily necessities. The government has designated 7,068 earthquake shelters nationwide, which can be located through the National Disaster Safety Portal and a smartphone application.",
"translated": "The citizens of Pohang witnessed the terrifying scene of apartment building walls cracking right before their eyes, leaving them shaken with fear throughout the day. Over 700 citizens are still unable to return home and are taking shelter elsewhere. We're going live to the scene. Reporter Ha Dong-won, can you tell us which shelter that is? Yes, it's the Heunghae Indoor Sports Center, which the city of Pohang has set up as a temporary evacuation center. Currently, around 500 people, including residents of Pohang's Buk-gu district, are taking refuge here. The 5.4-magnitude earthquake and ongoing aftershocks have continued to unsettle residents, with some areas experiencing cracks in apartment building walls, exposing internal steel frameworks and shattering entrance doors and windows, raising concerns about potential building collapses. The Pohang city disaster response headquarters has reported that over 730 people have been displaced. Many residents were forced to evacuate without even gathering essential items. Companies and organizations are providing emergency relief supplies, including water, ramen, and daily necessities. The National Safety Agency designated 7,068 earthquake shelters nationwide last December. The locations and status of these shelters can be found on the National Disaster Safety Portal and the Safety Dream DOL smartphone application. This is Ha Dong-won reporting live from Pohang, Gyeongbuk, for TV Chosun.",
"translated_llm": "The city of Port-au-Prince has been in a state of emergency since last week. The city has announced that more than 500 people have been evacuated from the area, including residents of Port-au-Prince, who were evacuated last week. The earthquake, which was 5.4 magnitude, has continued to cause residents to feel uneasy. In some areas, the outside of the building has been exposed to the outside and the inside of the building has been exposed to the railroad because of the crack in the building.",
"translated_whisper": " The citizens of Pohang were terrified by the sight of the apartment buildings being split in two. 700 citizens are still staying in the shelter without returning home. We'll get to the scene. Reporter Ha Dong-won, which shelter is that? Yes, it's the Henghae Indoor Sports Center, which Pohang has temporarily opened as a temporary shelter. It's in a state where 500 people including residents of Pohang were in the shelter. The 5.4 earthquake continued and the residents were not worried. Some areas were exposed to the outside of the apartment building, and the entrance window window was broken into pieces. The Ministry of Emergency Management said that there were 730 people. Most residents were not able to take their belongings. The company and the organization are supporting emergency relief supplies such as water, ramen, and daily necessities. The National Safety Agency has designated 7,068 earthquake shelters nationwide in December last year. The location and status of the earthquake shelters can be found in the National Disaster Safety Portal and the Safety Dream DOL smartphone application. This is Ha Dong-won from Gyeongbuk, Pohang, TV Chosun."
} |
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... | [
"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 road... | [
"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 Philip... | {
"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"
} | 32 | [
"In the frame, a person is seen assisting a dog with a red harness to board an airplane via a set of stairs. The airplane is prominently visible in the background, suggesting a preparation for a flight. The presence of a person and the dog implies a focus on animal transport or travel. The Korean text in the image ... | Be0D3w0LKSg | The video depicts the aftermath of Typhoon Namtheun's impact on Japan, showcasing severe weather conditions, flooding, and damage to infrastructure. The footage begins with a person assisting a dog onto an airplane, followed by scenes of preparation inside the airplane cabin. However, the majority of the video focuses on the typhoon's effects, including powerful waves crashing against coastal structures, flooding in residential areas, and strong winds causing damage. The storm's impact on daily life is highlighted, with scenes of flooded streets, submerged vehicles, and disrupted public transportation. The video also mentions Typhoon Songda, emphasizing the significant rainfall and severe impact on the local area. As the video progresses, it shifts to show the aftermath of the storm in South Korea, with scenes of flooded areas, damaged buildings, and people navigating through the challenging conditions. The video concludes with a call to action, encouraging viewers to subscribe to the channel, and ends with the News 1 TV branding, indicating the source of the news segment. Overall, the video provides a comprehensive look at the devastating effects of the typhoons on Japan and South Korea, highlighting the need for urgent action and response to such severe weather events. | {
"original": " Teksting av Nicolai Winther",
"refined": "Thank you for watching Nicolai Winther.",
"summarized": "Not Available",
"translated": "Text by Nicolai Winther",
"translated_llm": "Text by Nicolai Winther",
"translated_whisper": " Thanks for watching!"
} |
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... | [
"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 road... | [
"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 Philip... | {
"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"
} | 32 | [
"A weather presenter is reporting on Typhoon Nammador, emphasizing its impact on the Gyeongsang region and Gangwon's Yeongdong area, where strong winds and heavy rainfall are expected. The presenter discusses the typhoon's current position and forecasted path, highlighting that it is approaching the southeastern co... | AXoJU3hRho0 | The video describes the impact of Typhoon Nammador on South Korea, particularly the Gyeongsang region and Gangwon's Yeongdong area. The typhoon, the 14th of the season, is forecasted to move north at a speed of 17 km/h, making landfall near the southeastern coast of Busan and then approaching an area east-southeast of Ulsan. The expected rainfall is significant, with some areas receiving up to 150 mm, leading to potential flooding. The typhoon's influence is widespread, with strong winds and heavy rainfall affecting various regions, including the coastal areas of Gyeongsang, Gangwon's Yeongdong, and Ulleungdo/Dokdo. Following the typhoon's passage, the weather is expected to clear up, with a temperature drop of 5 to 10 degrees on the following day. The video also shows scenes of heavy rainfall, strong winds, and flooding, emphasizing the severe weather conditions caused by the typhoon. Overall, the video provides a detailed forecast and update on the typhoon's trajectory, intensity, and expected impact on South Korea. | {
"original": " 월요일인 오늘 태풍 난마돌의 영향으로 경상권과 강원 영동지역에 강한 비바람이 불겠습니다. 태풍 14호 난마돌은 오늘 오전 3시 현재 일본 가고시마 북북석쪽 약 140km 해상에서 시속 17km로 북진할 것으로 예측되는데요. 오늘 오전 9시 세력이 강상태로 다소 약화되며 부산 남동쪽 약 220km 부근 육상으로 진출 후 이날 오후 3시 울산 동남동쪽 약 240km 부근 육상으로 다가설 것으로 보입니다. 우리나라 주요 지점과 태풍 중심의 거리가 서귀포 440km, 통영 440km, 부산 450km, 울산 490km, 포항 540km로 태풍 영향권 주변으로 관측됩니다. 오늘 새벽까지 곳에 따라 시간당 최대 60mm의 폭우가 쏟아지는 곳도 있고, 예상 강수량은 경상권 해안, 광원 용동, 울릉도 독도는 50에서 100mm가 되겠습니다. 특히 경상권 해안에는 최고 150mm 이상의 물폭탄이 쏟아지겠으며 하늘은 대체로 흐리겠지만 저녁 들어 서쪽 지역부터 살짝 맑아질 것으로 보입니다. 태풍이 지난 후 화요일인 내일은 오늘보다 5도에서 10도가량 기온이 더 떨어질 것으로 보이며 기상청 중기 예보에 따르면 일요일인 오늘 25일까지 아침 최저기온은 14도에서 17도, 낮 최고기온은 23도에서 25도로 전망됩니다. 본격적인 신선한 가을 날씨가 당분간 이어질 것으로 보입니다",
"refined": "Today is Monday. Due to the influence of Typhoon Nammador, strong rain and wind will be expected in the Gyeongsang region and Gangwon's Yeongdong area. As of 3 am this morning, Typhoon Nammador No. 14 is predicted to move north at 17 km per hour, approximately 140 km north of Japan's Kagoshima. By 9 am, its power is expected to slightly weaken, and it will make landfall near Busan's southeast coast, around 220 km away, before approaching about 240 km southeast of Ulsan by 3 pm. The distance between major points in Korea and the typhoon's center is as follows: Seogwipo 440 km, Tongyeong 440 km, Busan 450 km, Ulsan 490 km, and Pohang 540 km, all within the typhoon's sphere of influence. Heavy rainfall has been observed in some areas, with hourly precipitation reaching up to 60 mm. The expected rainfall is 50-100 mm in the coastal areas of Gyeongsang-do, Gwangwon-dong, and Dokdo, with the Gyeongsang-do coast potentially experiencing over 150 mm. The sky will be mostly cloudy, but it is expected to clear up in the west by evening. Following the typhoon, tomorrow's temperature is expected to drop by 5-10 degrees. According to the mid-term weather forecast, the morning temperature on Sunday will be 14-17 degrees Celsius, while the daytime temperature will be 23-25 degrees Celsius, indicating a fresh autumn weather that is likely to continue for the time being.",
"summarized": "Typhoon Nammador is expected to bring strong rainstorms to the Gyeongsang and Gangwon regions in South Korea. The typhoon is currently 140km north of Japan's Kagoshima island and is expected to move north at 17km per hour. The storm's center is predicted to pass near Busan and Ulsan, with heavy rainfall expected in the Gyeongsang coast, Gwangwon, and Dokdo, with some areas receiving over 150mm of rain. The weather is expected to clear up by the evening, with a drop in temperature of 5-10 degrees on Tuesday. The autumn weather is forecast to continue, with morning temperatures ranging from 14-17°C and daytime temperatures ranging from 23-25°C.",
"translated": "Today, Monday, the Gyeongsang region and Gangwon's Yeongdong area will experience strong winds and heavy rainfall due to the influence of Typhoon Nammador. As of 3 am this morning, Typhoon Nammador, the 14th of the season, is expected to move north at a speed of 17 km/h, approximately 140 km northeast of Kagoshima, Japan. By 9 am, its strength will slightly weaken, and it is forecasted to make landfall near the southeastern coast of Busan, about 220 km away, and then approach an area approximately 240 km east-southeast of Ulsan by 3 pm today. The distance between major points in Korea and the center of the typhoon is approximately 440 km from Seogwipo, 440 km from Tongyeong, 450 km from Busan, 490 km from Ulsan, and 540 km from Pohang, indicating that these areas are on the periphery of the typhoon's influence. Until this morning, some areas experienced heavy rainfall of up to 60 mm per hour. The expected rainfall is 50-100 mm in the coastal areas of Gyeongsang, Gangwon's Yeongdong, and Ulleungdo/Dokdo, with the Gyeongsang coast potentially receiving over 150 mm, leading to significant flooding. The skies will be mostly cloudy, but are expected to clear up slightly in the western regions from the evening. Following the typhoon's passage, tomorrow, Tuesday, is expected to be 5 to 10 degrees cooler than today. According to the Korea Meteorological Administration's mid-term forecast, from today, the 25th, until Sunday, morning temperatures will range from 14 to 17 degrees Celsius, and daytime temperatures will range from 23 to 25 degrees Celsius, indicating a continuation of crisp autumn weather for the time being.",
"translated_llm": "On Monday, the typhoon reached the north-eastern part of the country and the south-eastern part of the country. The typhoon was expected to blow at about 14 km. The typhoon was expected to blow at about 140 km from the northern part of Japan. By 9 am, the force was weakening and the volcano was expected to reach the southeastern part of the country. By 3 pm, the typhoon was expected to reach the south-eastern part of the country.",
"translated_whisper": " Today is Monday. Due to the typhoon Nammador, strong rainstorms will be blowing in Gyeongsang-gu and Gangwon-dong. Typhoon Nammador No. 14 is expected to head north at a speed of 17km per hour at about 140km from the northern island of Kagoshima, Japan, at 3 a.m. this morning. At 9 a.m. this morning, the power is slightly weakened in strong condition, and it will advance to the north of 220km in the southeast of Busan, will approach the North Sea at 3 p.m. on the afternoon of this day, about 240 km north of Ulsan Dongnam-dong. The main point of the Korean island and the center of the typhoon are 440 km from the West Gipo, 440 km from the Tongyeong, 450 km from Busan, 490 km from Ulsan, and 540 km from Pohang. It is observed around the typhoon territory. There are places where the typhoon is pouring down at a maximum of 60 mm per hour, The expected rainfall will be 50 to 100 mm in Gyeongsang-do, Gwangwon-dong, and Dokdo. In particular, the Gyeongsang-do coast will be flooded with more than 150 mm of water, and the sky will be cloudy, but it will seem to clear up from the west side in the evening. Tomorrow, Tuesday, will be 5 to 10 degrees less than today. According to the weather forecast, and Sunday, today, 25th, the morning temperature is 14 to 17 degrees Celsius, and the day temperature is 23 to 25 degrees Celsius. It seems that the fresh autumn weather will continue for the time being."
} |
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... | [
"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 road... | [
"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 Philip... | {
"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"
} | 32 | [
"The current frame shows images of police officers in riot gear, equipped with shields and firearms. The context indicates the scene is related to a typhoon warning for the southeastern coastal areas of Korea, including Busan and Ulsan, due to Typhoon Nanmadol making landfall in Japan. The visible text on the image... | MYkG8SmCWNo | The video depicts a news broadcast covering Typhoon Nanmadol, a powerful storm that has made landfall in Japan. The broadcast shows images of police officers in riot gear preparing for the typhoon's impact, as well as satellite views of the storm from the International Space Station. A NASA astronaut, Bob "Farmer" Hines, is featured in a tweet expressing his prayers for the safety of those in the typhoon's path, along with photos of the storm taken from space. The broadcast emphasizes the severity of the typhoon, with warnings issued for southeastern coastal areas of Korea, including Busan and Ulsan. The news anchor delivers updates on the typhoon's progress, highlighting its immense size and potential impact. The video also includes images of the typhoon's eye and cloud patterns, as well as mentions of the United Nations World Meteorological Organization's release of images and a weather forecast. Throughout the broadcast, the importance of safety and preparedness is stressed, with encouragement for viewers to take necessary precautions and stay vigilant. | {
"original": " 지금까지 생생 지구촌이었습니다. 우주정거장에서 찍은 남마돌의 사진 3장을 올리며 태풍 남마동 경로에 있는 모든 사람들의 안전을 위해 기도한다고 적었습니다. 현재 초강력 태풍 남마돌은 일본에 상륙했고요. 우리나라 부산과 울산 등 영남 해안에도 태풍 경보가 내려진 상황입니다.",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "That's all for today's live coverage of our global village. A NASA astronaut posted three photos of Typhoon Nanmadol taken from the International Space Station and expressed his prayers for the safety of everyone in the typhoon's path. The extremely powerful Typhoon Nanmadol has currently made landfall in Japan, and a typhoon warning has been issued for the southeastern coastal areas of Korea, including Busan and Ulsan.",
"translated_llm": "Until now, it was a living earth village. He posted three photos of the southern storm from the space station and said he prayed for the safety of all people on the southern storm route.",
"translated_whisper": " The world of the eye. The 14th typhoon Nammadol, which was spotted at the International Space Station, has been revealed. The eyes of the typhoon are clear in the middle of a huge cloud of clouds. These are the pictures that Bob Hines, a NASA astronaut from the US Air Force, released on Twitter on the 18th. He wrote that he was praying for the safety of all people on the Taepung Namma-dong route, with three photos of the Namma-dol taken at the International Space Station. Currently, the ultra-powerful Taepung Namma-dol has landed in Japan. A typhoon warning has been sent to the southwest coast of Yeongnam, including Busan and Ulsan."
} |
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... | [
"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 road... | [
"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 Philip... | {
"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"
} | 32 | [
"The image depicts a news broadcast focused on the weather, specifically about Typhoon Nanmadol. The two anchors are seated at a news desk with weather maps displayed in the background. The map highlights the current position and projected path of the typhoon. The discussion involves the typhoon's trajectory, with ... | waVeQcHP3f0 | The video is a news broadcast about Typhoon Nanmadol, the 14th typhoon of the season, which is expected to affect South Korea, particularly Jeju Island and coastal areas. The typhoon's trajectory has shifted slightly more eastward, moving towards Japan, but its immense strength still poses concerns for strong winds and heavy rainfall in the region. The Korea Meteorological Administration has issued warnings for gusts of up to 35 meters per second in Jeju Island, the Yeongnam coast, and Ulleungdo and Dokdo islands, as well as heavy rainfall of up to 150mm in the Yeongnam coast. The critical period for the typhoon is from tonight to tomorrow morning, with potential storm surges and strong winds expected in Jeju Island and coastal areas. The news anchors and reporters emphasize the need for thorough preparations and caution in the affected regions, providing updates on the typhoon's path, wind speeds, and rainfall expectations. The video also shows footage of coastal areas, roads, and streets affected by the typhoon, highlighting the severe weather conditions and the impact on daily life. Overall, the video aims to inform and warn viewers about the typhoon's potential dangers and encourage them to take necessary precautions. | {
"original": " 북상 중인 14호 태풍 난마돌의 진로가 동쪽으로 조금 더 치우치는 것으로 분석됐습니다. 하지만 태풍의 세력이 워낙 강해 제주도와 해안가에는 여전히 비바람 피해가 우려됩니다. 취재기자 연결해 자세한 태풍 상황 알아보겠습니다. 김진두 기자. 네, 김진두입니다. 태풍이 당초 이상보다 일본 쪽으로 더 치우쳐서 이동한다는 건데 그래도 강한 비바람은 그대로 유지가 된다는 거죠? 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 비바람의 예보는 그대로 유지가 됐습니다. 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 초속 35m에 달하는 돌풍이 불겠다고 밝혔습니다. 그 밖의 해안과 경남 내륙에도 최고 초속 25m, 내륙지방에도 초속 15m의 순간적인 강풍이 예상됩니다. 비는 영남 해안과 영동 지방 그리고 울릉도 독도에 50에서 100mm가 오겠고 특히 영남 해안에는 150mm가 넘는 호우가 쏟아질 것으로 보입니다. 기상청은 이번 태풍의 고비가 오늘밤부터 내일 오전 사이가 되겠고 제주도와 해안가에는 폭풍의 일이나 월파 가능성도 있다며 철저한 대비를 당부했습니다. 지금까지 YTN 김진우입니다.",
"refined": "Typhoon Nammadole, currently moving north, is being analyzed as shifting slightly more to the east. Although its trajectory is leaning more towards Japan than initially thought, strong winds and rain are still expected to affect Jeju Island and the coast. Let's connect with our correspondent for more details. Reporter Kim Jin-du, can you tell us more? Yes, I'm Kim Jin-du. The typhoon is moving more towards Japan, but the strong winds and rain are expected to remain the same, right? That's correct. The weather agency announced in a morning briefing that the typhoon's route has been slightly adjusted. Although the typhoon is expected to make landfall in northern Kyushu, Japan, around tomorrow morning, its route will shift slightly more to the east. However, this doesn't change the forecast for strong winds and rain, as the typhoon is maintaining its powerful force. We can expect extremely strong winds, with concentrated heavy rain forecasted for the Yeongnam coast and Yeongdong region. According to the weather agency, Jeju Island, the Yeongnam coast, and Ulleungdo Dokdo can expect momentary gusts of up to 35 meters per second. Other coastal areas, including the mainland of Gyeongsangnam-do, may experience gusts of up to 25 meters per second, while inland areas may see momentary strong winds of up to 15 meters per second. Rainfall is expected to reach 50-100mm in the Yeongnam coast, Yeongdong region, and Ulleungdo Dokdo, with some areas in the Yeongnam coast potentially receiving over 150mm. The weather agency has warned that the critical period for the typhoon will be from tonight to tomorrow morning and has advised thorough preparations, as there is a possibility of storm surges or high waves in Jeju Island and coastal areas.",
"summarized": "Typhoon Nammadole, the 14th of the season, has shifted its trajectory slightly to the east but is still expected to bring strong winds and rain to Jeju Island and the coast. The Korea Meteorological Administration reported that the typhoon will land in northern Kyushu, Japan tomorrow morning. Despite the slight change in trajectory, the forecast for strong winds and rain remains the same. Wind gusts of up to 35 meters per second are expected in Jeju Island, the Yeongnam coast, and Ulleungdo, with heavy rain of 50-100mm predicted in these areas, and over 150mm on the Yeongnam coast. The weather service has warned of a high risk of storm damage and flooding, particularly in Jeju Island and coastal areas, and has advised thorough preparations.",
"translated": "The trajectory of Typhoon Nanmadol, the 14th typhoon of the season, has been analyzed to be shifting slightly more eastward. However, due to the storm's immense strength, concerns of wind and rain damage still persist, particularly in Jeju Island and coastal areas. Let's connect with our reporter for a more detailed update on the typhoon situation. Reporter Kim Jin-do, can you tell us more? Yes, I'm Kim Jin-do. The typhoon is now expected to move more towards Japan than initially predicted, but the strong winds will remain unchanged, correct? Yes, that's right. The Korea Meteorological Administration announced in a morning briefing that the typhoon's path has been slightly adjusted. While the typhoon is still expected to make landfall in northern Kyushu, Japan, tomorrow morning, its trajectory will shift slightly more eastward. Although this means the typhoon will be farther away from Korea, the forecast for strong winds and rain remains unchanged due to the storm's extreme strength. The winds will be extremely strong, with concentrated heavy rainfall expected in the Yeongnam coast and Yeongdong regions. The Korea Meteorological Administration warned that gusts of up to 35 meters per second can be expected in Jeju Island, the Yeongnam coast, and Ulleungdo and Dokdo islands. Additionally, coastal areas and the inland regions of Gyeongsang Province can expect gusts of up to 25 meters per second, while inland areas may experience gusts of up to 15 meters per second. Rainfall is expected to be between 50-100mm in the Yeongnam coast, Yeongdong, and Ulleungdo, with the Yeongnam coast potentially receiving over 150mm of rain. The Korea Meteorological Administration stated that the critical period for the typhoon will be from tonight to tomorrow morning and warned of potential storm surges and strong winds in Jeju Island and coastal areas, urging thorough preparations. This is YTN Kim Jin-woo.",
"translated_llm": "The 14th North China Typhoon Nanmadol is expected to move slightly more eastward, but the storm's strength is still very strong and there is still concern about wind damage on the coast. We will check with the media to see more details about the storm situation. Kim Jin-do reporter. Yes, Kim Jin-do appears. The storm is moving more strongly toward Japan than it did at first, but the winds are still strong. Yes. The weather bureau said in a morning briefing that the storm's path has been carefully adjusted.",
"translated_whisper": " The 14th typhoon Nammadole, which is in the north, is being analyzed as a little more The typhoon is moving more to Japan than it was at the time, but the strong rain and wind will remain the same, right? Yes, that's right. The weather office said that the typhoon's direction was finely adjusted in the briefing this morning. The typhoon will land in Kyushu, North Japan tomorrow morning, but the route will move a little more east. The distance from Korea is getting farther, but the forecast for rain and wind has remained the same. The wind is very strong and the wind is very strong. First of all, the wind is very strong. The Yeongnam coast and Yeongdong are expected to be concentrated. The weather office said that the maximum wind speed of the moment in Jeju Island, Yeongnam coast and Ulleungdo Dokdo is 35 meters per hour. In addition, the coast and the mainland of Gyeongnam are expected to have a maximum speed of 25 meters per hour. In the mainland, a momentary strong wind of 15 meters per hour is expected. Rain will be 50-100mm in the Yeongnam coast, Yeongdong province, and Ulleungdo province. In particular, the Yeongnam coast will be flooded with over 150mm of rain. The weather service said that the typhoon's high tide will be between tomorrow morning and the night of the day, and there is a possibility of a typhoon or a month-long wave in Jeju Island and the coast. This is YTN Kim Ji-do."
} |
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... | [
"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 road... | [
"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 Philip... | {
"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"
} | 32 | [
"The video is a news report discussing the impact of Typhoon Nanmadol, which has made landfall in Japan. The current frame shows a news anchor presenting information about the typhoon, highlighting that it is affecting Japan, South Korea's southern and eastern coasts, and the Yeongnam coast. The text on the screen ... | umbAs-igFhc | The video is a news report covering the impact of Typhoon Nanmadol, which made landfall in Japan and affected South Korea's southern and eastern coasts, including the Yeongnam region. The report details the typhoon's strength, with strong winds of up to 40 meters per second and a wide radius of approximately 400 kilometers, prompting the issuance of typhoon warnings and landslide warnings for areas such as Pohang and Gyeongju. The video shows nighttime scenes of affected areas, including wet roads, low visibility, and damage to buildings and infrastructure. Emergency responders, including firefighters, are depicted working to secure or repair structures, highlighting the ongoing efforts to manage the aftermath of the typhoon. The report emphasizes the need for preparedness and caution due to the severe weather conditions, including strong winds, high rainfall, and potential storm surges or high tides along the coast. As the typhoon weakens somewhat, the video continues to show the damage caused by the storm, with debris scattered on sidewalks, broken tiles, and damaged walls. The overall context is one of severe weather conditions and an active emergency response to mitigate the effects of Typhoon Nanmadol. | {
"original": " 태풍 난마돌과 관련해 방금 들어온 속보 전해드리고 뉴스 이어가겠습니다. 역대급 세력을 지닌 제14호 태풍 난마돌이 일본에 상륙하면서 우리나라 남해와 동해상은 물론 영남 해안에도 태풍경보가 내려졌습니다. 현재 부산과 울산 등 태풍특보 지역에는 초속 30m 안팎의 돌풍이 불고 있어 강풍 피해가 우려됩니다. 또 포항과 경조에는 산사태주의보가 발령 중입니다. 기상청은 14호 태풍 난마돌이 어제저녁 일본 가고시마 남쪽에 상륙한 뒤 태풍의 세력이 매우 강해서 강한 강도로 한 단계 약화했다고 밝혔습니다. 하지만 여전히 중심 기압은 955헥토파스칼로 강한 세력을 유지하고 있고 중심 부근에는 초속 40m의 돌풍이 몰아치고 있습니다. 특히 강풍 반경이 400km로 매우 넓어서 우리나라 남해와 동해상은 물론 부산과 울산, 호항, 경주 등 영남 해안에도 태풍 경보가 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 폭풍 해일의 가능성이 크다며 철저한 대비를 당부했습니다.",
"refined": "I'll now provide the latest news about Typhoon Nammador. The 14th typhoon, which possesses record-breaking strength, has made landfall in Japan. As a result, a typhoon warning has been issued for the southern and eastern coasts of Korea, as well as the Yeongnam coast. Currently, the typhoon alert areas, including Busan and Ulsan, are experiencing strong gusts of wind at around 30 meters per second, raising concerns about damage from strong winds. Additionally, a landslide warning has been issued for Pohang and Gyeongju. The Korea Meteorological Administration reported that after Typhoon Nammador made landfall south of Kagoshima, Japan, last night, its strength was significantly weakened by one level. Nevertheless, the central pressure remains at 955 hectopascals, indicating a powerful storm, with gusts of up to 40 meters per second near the center. Notably, the radius of the strong winds is approximately 400 kilometers, affecting not only the southern and eastern coasts of Korea but also the Yeongnam coast, including Busan, Ulsan, Pohang, and Gyeongju. The weather agency forecasts that the Yeongnam coast will experience gusts of 25 to 35 meters per second today, with up to 100mm of rainfall expected in Yeongnam, Yeongdong, Ulleungdo, and Dokdo. Moreover, waves of up to 10 meters high are expected in the sea, with a high risk of coastal flooding or storm surges, prompting a call for thorough preparations.",
"summarized": "Typhoon Nammador, the 14th of the season, has made landfall in Japan and is bringing strong winds and heavy rain to South Korea. A typhoon warning has been issued for the south and east coasts of the country, including the cities of Busan, Ulsan, Pohang, and Gyeongju. The storm is expected to bring winds of 25-35 meters per second to the Yeongnam coast and up to 100mm of rain to the Yeongnam, Yeongdong, and Ulleungdo Dokdo regions. The Korea Meteorological Administration has warned of a high possibility of storm surges and flooding along the coast, with waves reaching up to 10 meters high.",
"translated": "We have just received a breaking news report regarding Typhoon Nanmadol, and we will continue with the news. The 14th typhoon, Nanmadol, which boasts unprecedented strength, has made landfall in Japan, prompting the issuance of a typhoon warning for the southern and eastern coasts of South Korea, as well as the Yeongnam coast. Currently, areas under typhoon alerts, including Busan and Ulsan, are experiencing strong gusts of wind at around 30 meters per second, raising concerns about potential damage from strong winds. Additionally, landslide warnings have been issued for Pohang and Gyeongju. The Korea Meteorological Administration announced that Typhoon Nanmadol, which made landfall south of Kagoshima, Japan last night, had extremely strong winds, resulting in a one-level downgrade in its classification as a strong typhoon. Nevertheless, the storm still maintains a central pressure of 955 hectopascals and is accompanied by powerful gusts of up to 40 meters per second near its center. Notably, the radius of the strong winds is approximately 400 kilometers, which is unusually wide, leading to the issuance of typhoon warnings for the southern and eastern coasts of South Korea, including Busan, Ulsan, Pohang, and Gyeongju. The weather agency forecasted that the Yeongnam coast will experience strong gusts of 25-35 meters per second today, with expected rainfall of up to 100mm in the Yeongnam, Yeongdong, Ulleungdo, and Dokdo regions. Furthermore, the agency warned of high waves reaching up to 10 meters in the sea and a significant risk of storm surges or high tides along the coast, emphasizing the need for thorough preparations.",
"translated_llm": "The 14th Typhoon, with a maximum force of 14 knots, has landed in Japan and has been downgraded to a very strong typhoon. However, the central pressure is still very strong and the winds are likely to remain strong at 955 hectolitres and the winds are likely to be strong and blowing at 40mph in the central region. The typhoon is expected to be very strong, especially in the southern part of the country.",
"translated_whisper": " I'll tell you the news about the typhoon Nammador. The 14th typhoon Nammador, which has the highest power, landed in Japan. The typhoon warning was also sent to the south and east coast of Korea, as well as the south-south coast. Currently, typhoon special forces in Busan and Ulsan are in a storm at an early 30m. There is a risk of strong wind damage. Also, the mountain situation warning is being issued in Pohang and Gyeongju. The Air Force said that the 14th Typhoon Namba was landed in the south of Kagoshima, Japan last night, and the force of the typhoon was very strong, so it was weakened by a step. However, the central pressure is still maintained at 955 hectopascal, and a 40-meter gust of wind is blowing around the center. In particular, the radius of the wind is very wide at 400 kilometers, so typhoons are being launched on the south and east coast of South Korea, and a typhoon warning is being issued on the Yeongnam coast, including Busan, Ulsan, and Hohang, Gyeongju. The weather office said that a gust of wind of 25 to 35 meters will blow at the Yeongnam coast today, and a maximum of 100mm of rain is expected in Yeongnam, Yeongdong, and Ulleungdo Dokdo. In particular, the sea level is expected to be 10 meters high, and there is a high possibility of a monthly wave or a storm at the coast."
} |
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... | [
"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 road... | [
"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 Philip... | {
"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"
} | 32 | [
"A news anchor is standing in front of a background depicting the weather system of Typhoon Nanmadol. The background shows a map with Japan's outline and a visual representation of the storm. The news anchor appears to be discussing the typhoon's movement and the warnings issued by the Japan Meteorological Agency a... | fxLzed6AI2Q | The video depicts the impact of Typhoon Nanmadol, which made landfall in the southern part of Kyushu, Japan. The typhoon brought record-breaking strong winds and heavy rain, causing widespread damage and destruction to buildings, infrastructure, and the environment. The Japan Meteorological Agency warned of an unprecedentedly dangerous storm with the potential for a large-scale disaster, urging special precautions. The video shows scenes of heavy rain, strong winds, and low visibility, as well as damaged buildings, debris-filled streets, and downed power lines. The typhoon's path was predicted to move northward along the western inland area of Kyushu, then change direction to the northeast, crossing the Japanese archipelago and moving out to sea east of Hokkaido. The video highlights the severe impact of the typhoon on the region, with scenes of destruction, debris, and hazardous conditions, emphasizing the need for caution and the potential for large-scale disasters. | {
"original": " 14호 태풍 난마돌이 일본 규슈 남부에 상륙했습니다. 일본 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, 포트카이도 동쪽 해역으로 빠져나갈 전망입니다. 일본 기상청은 이번 태풍이 지금까지 경험한 적 없는 위험한 태풍으로 수십 년에 한 번 있을 만한 대규모 재해가 발생할 가능성이 있다며 각별한 대비를 당부했습니다.",
"refined": "The 14th typhoon, Nanmadol, has landed in southern Kyushu, Japan. The Japan Meteorological Agency predicts that the typhoon will move northward along western Kyushu and then change direction to the northeast in northern Kyushu this morning. The typhoon is expected to traverse the Japanese archipelago and exit into the sea east of Hokkaido on the afternoon of the 21st. The Japan Meteorological Agency warns that this typhoon is an unprecedentedly dangerous one, which may bring a large-scale disaster that occurs only once in decades, and is urging extra precautions.",
"summarized": "Typhoon Namba has landed in southern Kyushu, Japan, and is expected to move northward through western Kyushu and then northeast from northern Kyushu. It is predicted to pass through Japan and exit into the sea east of Hokkaido by the afternoon of the 21st. The Japan Meteorological Service has warned of a high risk of a large-scale disaster, describing the typhoon as unprecedented and potentially catastrophic, urging special precautions.",
"translated": "The 14th typhoon, Nanmadol, has made landfall in the southern part of Kyushu, Japan. The Japan Meteorological Agency predicts that the typhoon will move northward along the western inland area of Kyushu and then change direction to the northeast from northern Kyushu today. The typhoon is expected to cross the Japanese archipelago and move out into the sea east of Hokkaido by the afternoon of the 21st. The Japan Meteorological Agency has warned that this typhoon is an unprecedentedly dangerous storm, with a possibility of a large-scale disaster occurring that happens only once every few decades, and has urged special precautions.",
"translated_llm": "The 14th Typhoon Nanma Dō landed in the southern part of Kyūshū. The Japan Meteorological Agency forecast that this typhoon would move north-eastward from Kyūshū in the northern part of the country today. The typhoon will cross the Japanese mainland on the afternoon of the 21st, and will also leave Port Kaido in the eastern seaboard.",
"translated_whisper": " The 14th typhoon Namba hit the southern part of Kyushu, Japan. The Japan Meteorological Service predicted that the typhoon would follow the western part of Kyushu and move northward from Kyushu to the northeast this morning. The typhoon will penetrate Japan's tropical heat and escape to the east of Hokkaido on the afternoon of the 21st. The Japan Space Agency said that this typhoon is a dangerous typhoon that has never been experienced before, and that there is a possibility of large-scale disaster that may occur once every decade."
} |
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... | [
"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 road... | [
"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 Philip... | {
"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"
} | 32 | [
"In the current frame, a news reporter is delivering a weather update in an urban outdoor setting. The reporter is holding a microphone with the YTN logo and is wearing a purple dress. The background features trees and a building structure, indicating she is standing in a city park or a similar area. The visible te... | tVUSMdbETuE | A news reporter from YTN delivers a weather update from an urban outdoor setting, discussing the indirect influence of Typhoon Muifa, which is heading towards China, and the formation of the 14th typhoon of the season, Nammador, in the eastern waters of Okinawa, Japan. The reporter provides details on the expected weather conditions, including light rain in western coastal areas until morning, late summer heat with temperatures rising to 28 degrees in Seoul and 31 degrees in Gwangju, and a forecast of continued heat over the weekend with potential rain in Jeju Island's eastern parts. The typhoon's potential path and impact are discussed, with a strong wind warning issued for Jeju Island and the southern coast, and a high wave warning for most coastal areas. The reporter advises caution to prevent coastal accidents due to strong winds and high waves. The video also shows scenes of people engaging in daily activities despite the weather conditions, emphasizing the need to stay updated on weather forecasts amidst potential typhoon-related disruptions. | {
"original": " 오늘도 맑은 하늘 보기는 어렵겠습니다. 중국으로 향하는 태풍 무이파의 간접 영향으로 오전까지 중부 서해안에는 약한 비가 내리겠습니다. 비의 양이 워낙 적어서 낮 동안 활동하시는데 큰 불편은 없겠습니다. 때아닌 늦더위도 계속되겠습니다. 오늘 서울이 한낮에 28도, 광주가 31도까지 오르는 등 어제보다 덥겠습니다. 서울 등 수도권은 낮부터 비가 그치고 날이 점차 개겠습니다. 오늘 낮 기온은 예년보다 1에서 4도가량 높겠고 특히 호담 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 난마돌이 발생했습니다. 이 태풍은 세력이 강해지면서 느리게 북상해 다음 주 월요일 서귀포 남동쪽 약 280km 부근 해상까지 올라오겠습니다. 이후 진로는 유동적이지만 일본 규슈 부근이나 대한해업으로 이동할 가능성이 있습니다. 앞으로 발표되는 태풍 정보를 잘 확인해 주시기 바랍니다. 현재 태풍 무이파의 간접 영향으로 제주도와 남해안에는 강풍주의보가, 대부분 해상에는 풍랑주의보가 발효 중입니다. 또 제주도와 동해안에는 앞으로 비가 더 내릴 가능성이 있는 만큼 해안간 안전사고에 각별히 유의하시기 바랍니다. 지금까지 서울 상암동에서 YTN 원이다입니다.",
"refined": "It will be difficult to see clear skies today. The indirect influence of Typhoon Muifa, which is heading to China, will bring light rain to the western coast of South Korea until morning. However, the rain will be minimal, and there will be no significant disruption to your daily activities. The late heat wave will continue, with temperatures in Seoul and Gwangju expected to reach 28 and 31 degrees, respectively, making it hotter than yesterday. From daytime onwards, the rain will clear, and the skies will gradually become clearer. The daytime temperature is expected to be 1 to 4 degrees higher than average, with a significant rise in the Ho-dam region. The forecasted temperatures are 28 degrees in Seoul, 30 degrees in Jeonju, and 31 degrees in Gwangju. This weekend, the late heat wave is expected to intensify, with daytime temperatures in Seoul reaching up to 30 degrees. Rain is also expected, mainly in Jeju Island and the eastern regions. Meanwhile, the 14th typhoon, Nammador, formed in the eastern waters of Okinawa, Japan, at around 3 am today. As it strengthens, the typhoon will move northward slowly, reaching waters approximately 280 km south of Seogwipo by Monday. Its future route is uncertain, but it may move towards Kyushu or the Korean Peninsula. Please check the upcoming typhoon reports for updates. Due to Typhoon Muifa's indirect influence, strong wind warnings are in effect for Jeju Island and the southern coast, while rough sea warnings are in effect for most of the coastal areas. Additionally, please exercise caution regarding coastal safety accidents, as there is a possibility of more rain in Jeju Island and the East Sea.",
"summarized": "A weak rain is expected in the western coast of South Korea until morning due to Typhoon Muifa's indirect influence, but it will clear up during the day with rising temperatures. Today's forecast includes highs of 28 degrees in Seoul, 30 degrees in Jeonju, and 31 degrees in Gwangju, 1-4 degrees higher than usual. The late heat wave is expected to continue through the weekend, with temperatures possibly reaching 30 degrees in Seoul. Meanwhile, Typhoon Nammador has formed in the eastern sea of Okinawa, Japan, and is expected to move north slowly, possibly affecting the Korean Peninsula or Japan's Kyushu region by Monday. Strong winds and rough seas are forecasted for Jeju Island and the south coast, with a risk of coastal safety accidents due to heavy rain.",
"translated": "It will be difficult to see clear skies today due to the indirect influence of Typhoon Muifa, which is heading towards China, bringing light rain to the western coastal areas until morning. However, the small amount of rainfall is unlikely to cause significant disruptions to your daily activities. Meanwhile, the late summer heat will persist, with temperatures expected to rise, reaching 28 degrees in Seoul and 31 degrees in Gwangju, making it hotter than yesterday. As the rain clears up in Seoul and the metropolitan area from the afternoon, the weather will gradually improve. Today's high temperature is expected to be 1 to 4 degrees higher than usual, with a more significant rise in temperature expected in the Ho-dam region. The forecasted high temperatures are 28 degrees in Seoul, 30 degrees in Jeonju, and 31 degrees in Gwangju. Looking ahead to the weekend, the late summer heat is expected to continue, with temperatures potentially reaching 30 degrees in Seoul, and rain is forecasted mainly in the eastern parts of Jeju Island. In other news, the 14th typhoon of the season, Nammador, formed in the eastern waters of Okinawa, Japan, around 3 am this morning. As it strengthens, the typhoon is expected to move slowly northward, potentially reaching the waters south of Seogwipo, approximately 280 km away, by next Monday. Although its future path is uncertain, it may move towards the Kyushu region of Japan or the Korean Peninsula. Please stay tuned for further updates on the typhoon. Currently, a strong wind warning has been issued for Jeju Island and the southern coast due to the indirect influence of Typhoon Muifa, while a high wave warning is in effect for most coastal areas. Additionally, there is a possibility of more rain in Jeju Island and the eastern coastal areas, so please exercise extra caution to prevent coastal accidents. This is YTN Wonida reporting from Seoul's Sangam-dong.",
"translated_llm": "The weather will be warmer than yesterday. The city of Seoul will be warmer than yesterday, with winds of 28 degrees during the day and 31 degrees in Guangzhou. The city of Seoul will be affected by winds of 3 degrees and winds of 14 degrees in the middle of the month. The winds of the month will move more slowly than the winds of the month.",
"translated_whisper": " It will be difficult to see clear skies today. The typhoon Muipa, which is heading to China, will be a weak rain in the western part of China until the morning. There will be no big inconvenience in your activities during the day because there is so little rain. The late heat will continue. Seoul will be 28 degrees in the afternoon and Gwangju will rise to 31 degrees, which will be hotter than yesterday. Seoul and other cities will be rainy from the daytime and the day will gradually change. The temperature will be 4 degrees higher than the previous year, and the temperature will rise mainly in the Ho-dam region. Seoul 28 degrees, Jeonju 30 degrees, and Gwangju 31 degrees are expected. On the coming weekend, the temperature will rise to 30 degrees in Seoul, and the weather will be late. It is expected to rain mainly in the east of Jeju Island. Meanwhile, at around 3 a.m. this morning, the 14th typhoon Nammador occurred in the eastern sea of Okinawa, Japan. The typhoon will slowly rise as it gets stronger and will reach the southern part of the island on Monday. The route will be flexible, but it is possible that it will move to Kyushu or the coast of the Korean Peninsula. Please check the typhoon information that will be released in the future. Due to the indirect impact of the typhoon, the wind and wind warnings are being provided in Jeju Island and the south coast. Also, it is possible that it will rain more in Jeju Island and the East Sea, so please be careful of coastal safety accidents. This is YTN WONIDA from Seoul Sangam-dong."
} |
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... | [
"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 road... | [
"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 Philip... | {
"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"
} | 32 | [
"The current frame shows palm trees swaying vigorously in strong winds, likely caused by Typhoon Nanmadol. The image suggests the intensity of the typhoon, with visible motion blur indicating high wind speeds. The text overlay \"태풍 '난바돌' 근접 日, 900만 명 대피 지시\" (Typhoon 'Nanmadol' approaches, Japan, 9 million people i... | d6fZyZvF88k | The video depicts the severe impact of Typhoon Nanmadol on Japan, particularly in the Kyushu region. The typhoon brought extremely strong winds, with gusts reaching up to 50 meters per second, causing widespread damage and injuries. The video shows palm trees swaying violently, buildings with damaged roofs, and flooded streets, highlighting the extent of the destruction. The typhoon resulted in dozens of injuries, with over 40 people reported injured, and significant property damage, including toppled construction cranes and torn-off roofs. Approximately 9.6 million residents in the affected areas were issued the highest level of disaster warning and evacuation orders, urging them to secure their safety promptly. The video captures the hazardous conditions, including flooded rivers, damaged structures, and debris-filled streets, emphasizing the need for caution and safety in the affected areas. Overall, the video showcases the devastating effects of Typhoon Nanmadol on Japan, underscoring the urgency for residents to take immediate action to protect themselves and their properties. | {
"original": " 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. 순간 최대 풍속 초속 50m에 이르는 강풍으로 공사용 대형 크레인이 꺾이고 지붕이 날아가는 등 물적 피해도 속출했습니다.",
"refined": "Last night, dozens of people were injured due to the impact of Typhoon Nanmadol, which landed in Kyushu, Japan. According to NHK, over 40 people were injured, including those who fell due to strong winds. As a result, nearly 9.6 million local residents in the affected areas were asked to quickly secure their safety, with the highest level of disaster warning and evacuation instructions issued. The maximum wind speed reached 50 meters per second, causing physical damage such as the collapse of large cranes and roofs being blown away.",
"summarized": "Typhoon Nanmadol made landfall in Kyushu, Japan, causing significant damage and injuries. Over 40 people were injured, including those hurt by strong winds, while nearly 9.6 million residents were issued the highest level of disaster warnings and evacuation orders. The storm brought powerful gusts of up to 50 meters per second, resulting in destruction such as a large crane being bent and roofs being blown away, as well as water damage.",
"translated": "The typhoon Nanmadol, which made landfall in Japan's Kyushu region last night, has resulted in dozens of injuries and significant damage. According to NHK, over 40 people have been injured so far, including those who were blown over by strong winds. In response, the highest level of disaster warning and evacuation orders have been issued to approximately 9.6 million residents in the affected areas, urging them to secure their safety promptly. The typhoon brought extremely strong winds, with gusts reaching up to 50 meters per second, causing a large construction crane to topple and roofs to be torn off, resulting in substantial property damage.",
"translated_llm": "The impact of the typhoon, which landed in Kyushu, Japan last night, caused dozens of injuries and damage. NHK reported that over 40 people were injured by the typhoon so far. Along with this, the highest level of disaster warning and evacuation orders were issued to the local population of about 9.6 million people in the typhoon affected area requesting quick safety.",
"translated_whisper": " Last night, dozens of people were injured in the impact of the typhoon Nannamado, which landed in Kyushu, Japan. NHK said that more than 40 people were injured, including those who fell into a strong wind. With this, the local residents in the typhoon-affected area were asked to quickly secure safety, and the highest level of disaster warning and evacuation instructions were issued. The wind was at a maximum speed of 50 meters per second, causing a large crane to be bent and the roof to fly away. It also caused water damage."
} |
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... | [
"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 road... | [
"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 Philip... | {
"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"
} | 32 | [
"The current frame shows a satellite image of the East Asian region, highlighting several typhoons near the Korean Peninsula and Japan. The image includes three large typhoons over the ocean, with the largest one positioned northeast of Japan. The text in Korean at the bottom of the image translates to \"Autumn typ... | ItDO3MSUQRQ | The video discusses the presence of multiple typhoons near the Korean Peninsula, specifically focusing on the 12th, 13th, and 14th typhoons: Muifa, Merbok, and Nanmadol. Muifa is moving near the eastern coast of China, indirectly affecting Jeju Island, while Merbok is not expected to impact Korea. The primary concern is Nanmadol, which is moving north from the southeastern sea of Okinawa, Japan, and is expected to gain strength over warm waters, potentially becoming a powerful storm. Its strength is likened to a strong river capable of derailing a train. Nanmadol's path is unpredictable, and it is forecasted to approach the southeastern sea off Jeju Island's Seogwipo by next Monday. The Korea Meteorological Administration urges the public to monitor the situation closely and check weather updates frequently due to the typhoon's potential to cause significant impact. The video depicts the stormy conditions, turbulent seas, and potential aftermath of the typhoon, emphasizing the need for caution and preparedness. | {
"original": " 한반도 근처에서 관측되고 있는 올 가을 태풍들입니다. 제12호 태풍 무이파는 중국 동쪽 해안으로 빠져 제주도에 간접 영향을 주고 있는 상황. 제13호 태풍 무르복은 일본 동해상 먼바다에서 이동 중이라 우리나라에 영향을 주진 않을 전망입니다. 문제는 일본 오키나와 남동쪽 해상에서 발생해 북상 중인 제14호 태풍 난마돌입니다. 미크로네시아 유명 유적지의 이름을 따온 난마돌은 따뜻한 바다를 지나며 세력을 키워 모래에는 강도강으로 성장할 걸로 보입니다 강도강은 기차를 탈선시킬 수 있을 정도의 위력입니다 다음 주 월요일에는 위력을 유지하며 우리나라 제주 서귀포 남동쪽 해상까지 올라올 걸로 예측됩니다 하지만 아직까지는 이동 경로가 유동적인 만큼 상황을 조금 더 지켜봐야 합니다 기상청은 남마돌이 태풍이 발달하기에 좋은 환경의 해역을 지나고 있다며 관련 기상 정보를 수시로 확인해달라고 당부했습니다.",
"refined": "These are the autumn typhoons observed near the Korean Peninsula. The 12th typhoon, Muifa, has moved to China's eastern coast, indirectly affecting Jeju Island. The 13th typhoon, Merbok, is moving across the East Sea of Japan and is not expected to affect Korea. However, the 14th typhoon, Nammadol, which originated southeast of Okinawa, Japan, is a concern. Named after a famous Micronesian site, Nammadol is strengthening as it passes over warm waters and is expected to become a strong storm. Next Monday, it is predicted to maintain its power and approach the southeastern sea off Jeju Island's Seogwipo. However, the storm's path is still uncertain, so the situation needs to be monitored further. The Korea Meteorological Administration advised checking related weather information frequently, as Nammadol is currently in an environment conducive to its development.",
"summarized": "There are three autumn typhoons observed near the Korean Peninsula. The 12th typhoon, Muifa, is affecting Jeju Island indirectly, while the 13th typhoon, Merbok, is unlikely to affect Korea. The 14th typhoon, Nanmadol, is the main concern, named after a famous Micronesian site, and is expected to gain strength as it passes through warm waters, potentially growing powerful enough to derail a train. It is forecasted to reach the seas south of Jeju Island by next Monday, but its path remains unpredictable. The Korea Meteorological Administration advises checking weather updates frequently as Nanmadol is developing in a favorable environment.",
"translated": "These are the autumn typhoons being observed near the Korean Peninsula. The 12th typhoon, Muifa, is moving near the eastern coast of China, indirectly affecting Jeju Island. The 13th typhoon, Merbok, is currently over the distant waters of the East Sea of Japan and is not expected to impact Korea. The concern is the 14th typhoon, Nanmadol, which is moving north from the southeastern sea of Okinawa, Japan. Named after a famous site in Micronesia, Nanmadol is expected to gain strength as it passes over warm waters and potentially develop into a powerful storm. Its strength could be equivalent to a strong river that can derail a train. By next Monday, it is predicted to maintain its strength and approach the southeastern sea off Jeju Island's Seogwipo, but its path remains unpredictable, so the situation needs to be monitored closely. The Korea Meteorological Administration has urged the public to check weather updates frequently, as Nanmadol is currently in an environment conducive to its development.",
"translated_llm": "The 12th Typhoon Muay Pa is moving from the East Sea of Japan to the Far East Sea of Japan and is expected to affect our country. The problem is the 14th Typhoon Muay Pa in the North Sea, which is occurring in Okinawa and the South East Sea of Japan. The name of the famous place in Micronesia is taken from the fact that the Muay Pa will cross the warm sea and grow into a sandy river.",
"translated_whisper": " These are the autumn typhoons observed near the Korean Peninsula. The 12th typhoon, Muipa, is in the eastern coast of China, affecting Jeju Island indirectly. The 13th typhoon, Murubok, is moving in the far sea of the Japan East Sea, so it is expected not to affect Korea. The problem is the 14th typhoon, Nammador, which is in the north of the Okinawa South East Sea in Japan. The Nammadol, which has been named the famous site of Micronesia, will grow up through the warm sea and grow into a river in the sand. The river is so powerful that it can even displace a train. Next Monday, it is predicted that it will maintain its power and come up to the South East Sea of Jeju, Seogwipo, Korea. But so far, the moving route is flexible, so we need to wait a little longer. It is expected to rise to the west of Kyushu in Japan in 19 days, but the impact of our country is large due to the strength and change of the surrounding climate. The Ministry of the Environment said that the weather service is undergoing a good environment for the South Madol typhoon to develop, and asked to check the related weather information frequently."
} |
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... | [
"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 road... | [
"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 Philip... | {
"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"
} | 32 | [
"A news anchor is presenting information about the 14th typhoon, Nammadol, in a studio setting. The background displays visuals related to weather and technology. The anchor discusses the forecasted path of Nammadol, noting that it is expected to land in Kyushu, Japan, after moving north from the sea southeast of S... | ZuMemCoZIUY | The video reports on the 14th typhoon, Nammadol, which is expected to make landfall in Kyushu, Japan. The typhoon's forecasted path indicates that it will move north to the sea southeast of Seogwipo before changing direction and landing in Kyushu around next Monday. Although the possibility of significant damage to Korea has decreased, regions such as Jeju Island, the southern coast, and the eastern coast of the Yeongnam area are expected to experience strong winds and rain due to the typhoon's influence. The Korea Meteorological Administration emphasizes the need for caution, as the typhoon's path is still subject to change, and urges the public to closely monitor the forecasted trajectory. The video shows various scenes of coastal areas, including rough seas, strong winds, and people preparing for the typhoon's impact, highlighting the importance of vigilance and preparedness in the face of the approaching storm. | {
"original": " 북상족인 제14호 태풍 남마돌이 일본 규슈에 상륙할 것으로 보입니다. 기상청은 남마돌이 서귀포 남동쪽 먼 바다까지 북상한 뒤 급격히 방향을 틀어서 다음주 월요일쯤 일본 규슈에 상륙할 것으로 전망했습니다. 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 발표돼 있는 태풍 진로에 주의를 기울여달라고 당부했습니다.",
"refined": "The 14th typhoon, Nammadori, is expected to make landfall in Japan's Kyushu region. After moving north to the seas southeast of Seogwipo, it will change direction sharply and make landfall in Kyushu around next Monday. It is then expected to continue moving northeast along the Japanese archipelago. As a result, the possibility of significant damage to Korea has decreased, but Jeju Island, the southern coast, and the eastern coast of the southeastern region are likely to experience strong winds and rain due to the typhoon's direct or indirect influence. The weather office warns that the typhoon's path remains unpredictable and advises careful attention to future updates.",
"summarized": "Typhoon Nammadori, the 14th of the season, is forecasted to make landfall in Japan's Kyushu region. The storm is expected to move north along the Japanese Peninsula. While the likelihood of significant damage to Korea has decreased, Jeju Island, the southern coast, and the southeastern coast of the Korean Peninsula may experience strong winds and rain due to the typhoon's indirect influence. The weather office advises caution as the typhoon's path remains unpredictable.",
"translated": "The 14th typhoon, Nammadol, is expected to make landfall in Kyushu, Japan. The Korea Meteorological Administration forecasts that after moving north to the sea southeast of Seogwipo, Nammadol will suddenly change direction and land in Kyushu, Japan around next Monday. It is then expected to continue moving northeast along the Japanese archipelago. As a result, the possibility of significant damage to Korea has decreased, but Jeju Island, the southern coast, and the eastern coast of the Yeongnam region are likely to experience strong winds and rain due to the direct and indirect influence of the typhoon. The Korea Meteorological Administration urged caution, noting that the typhoon's path is still subject to change, and asked the public to closely monitor the forecasted typhoon trajectory.",
"translated_llm": "The weather service forecast that the storm would land in Japan on Monday. The weather service was told that the storm would continue to move northward, but that the winds would be stronger on the southern coast of the country. The weather service was asked to pay attention to the storm's path as it is still moving forward.",
"translated_whisper": " The 14th typhoon, the Nammadori, is expected to land in Kyushu, Japan. The weather office expected the Nammadori to land in Kyushu, Japan, and the Nammadori to land in the far east of the South East Sea of Seogwipo. It is expected to continue to move north along the Japanese Peninsula. Therefore, the possibility of great damage to Korea has been lowered, but the Jejudo and South Sea and the East and South Sea of the South The weather forecast says the typhoon's course is still flexible, so it is requested to be careful of the future typhoon course."
} |
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, S... | [
"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 Distri... | [
"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 grou... | {
"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"
} | 32 | [
"In the current frame, the focus is on an indoor subway station, with water visibly leaking from the ceiling onto the platform. The water streams down, creating a misty effect on the floor. There are informational signs and maps on the walls, indicating the station's location and navigation details. The area appear... | 1556640821054787588 | The video depicts an indoor subway station experiencing significant water leakage from the ceiling, creating a misty and hazardous environment on the platform. The leakage persists throughout the video, with water streaming down and pooling on the floor, posing a safety concern for potential passengers. Informational signs and maps on the walls remain visible, providing navigation details about the station despite the ongoing water issue. The situation escalates as the water continues to pour down, worsening the conditions and heightening safety concerns. The video shows no resolution to the water leakage issue, leaving the station in a state of potential danger for anyone who might be present. | {
"original": " Hello!",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Hello!",
"translated_llm": "What's going on?",
"translated_whisper": " Welcome to..."
} |
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, S... | [
"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 Distri... | [
"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 grou... | {
"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"
} | 32 | [
"The current frame shows a street flooded with water during heavy rainfall. A person is seen walking through the water, surrounded by parked buses and other vehicles. The area is wet and muddy, with rain pouring down. The scene appears to be in an urban setting, likely a bus terminal or a station, as suggested by t... | 1556678909436932096 | A severe flooding event is depicted in an urban setting, likely a bus terminal or station, with intense rainfall causing the street to be heavily inundated with water. The flooding is so severe that it partially submerges buses, cars, and road elements like poles, creating a chaotic and hazardous environment. Despite the challenging conditions, vehicles, including buses and cars, are seen navigating through the turbulent waters, with traffic signals still operational. The situation is commented on in a humorous tone through Korean text overlays, expressing surprise and frustration at the unexpected severity of the flooding and the difficulties it poses to drivers. The flooding continues throughout the video, with the water level remaining high and the current strong, indicating a significant and ongoing impact on the city, particularly around the 사당역 (Sadang Station) neighborhood. | {
"original": "",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Not Available",
"translated_llm": "- What?",
"translated_whisper": ""
} |
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, S... | [
"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 Distri... | [
"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 grou... | {
"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"
} | 32 | [
"A person stands in a flooded street, holding an umbrella and looking at their phone. The water level is high enough to submerge parts of their legs, and debris floats on the surface. The street is lined with buildings displaying various signs in Korean, and traffic lights are visible in the background. The scene s... | 1556973165556903939 | The video depicts a series of scenes showing the impact of heavy rain on an urban environment, likely in Korea given the presence of Korean signs and text. The footage begins with images of flooded streets, where people are navigating through the water, some holding umbrellas, and cars are attempting to drive through the flooded areas. The scenes then transition to show people engaged in various activities such as skateboarding, managing floodwaters, and conducting maintenance or cleaning tasks, both indoors and outdoors.
As the video progresses, it shows a bustling urban setting with people walking, using public transportation, and going about their daily lives despite the rain. There are scenes of workers managing water flooding inside a subway station, using brooms, buckets, and sandbags to mitigate the situation. The footage also captures the resilience of city life, with shops and establishments remaining open and people adapting to the weather conditions.
Throughout the video, the rain persists, casting a reflective sheen on the wet streets and illuminating the neon signs and lights from the shops. The scenes continue to depict daily life under rainy conditions, with people walking, driving, and engaging in conversations, highlighting the adaptability and resilience of the urban environment.
The original ASR transcript does not directly relate to the visual content of the video, suggesting it may be unrelated or possibly a separate audio track. However, it mentions the importance of watching and tending to something, possibly referring to the need for maintenance and care in the face of challenging weather conditions, which aligns with some of the activities depicted in the video. Overall, the video presents a dynamic and resilient urban landscape coping with the challenges posed by heavy rainfall. | {
"original": " Diolch yn fawr i'ch gilydd, Gwynt. Mae'n rhoi cherdded i'ch ceanfa. Mae'r caebl yn ddechrau, bob nigh at bob un. Rhaid eu gwylio byth a'u llwylyd. Mae'n dda iawn i ni'r rhan dechrau. Mae wedi cyrraedd yn bydd. Mae'r rhan fwyaf o'r brin yn cael ei gael ar y gwaith. Mae'r brin yn cael ei gael ar y gwaith.",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Thank you very much to you all, Gwynt. It gives a boost to your gathering. The cable is starting, each one to each other. They must be watched always and tended. It's very good for us to start. It has arrived as it should. Most of the work is being done on the spot. The work is being done on the spot.",
"translated_llm": "Thank you very much, Gwynt. I've given you my heart. I've given you my heart. I've given you my heart. I've given you my heart. I've given you my heart.",
"translated_whisper": " It's given to us all The gap is round each night and each morn We watch it pass and grow It is all in our hands, it is all in our hands With every field and rising sea Can hear the sounds of old And with every change we'll always be Where hope's not lost but found It is all in our heads, it is all in our heads It is all or nothing I'm going to show you the process of making a"
} |
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, S... | [
"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 Distri... | [
"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 grou... | {
"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"
} | 32 | [
"The current frame depicts a nighttime scene of a flooded area, likely caused by heavy rain or a storm. Water is rushing through the street, creating turbulent waves and foam. The area is illuminated by streetlights, and the vegetation on both sides of the road is visible. The railing in the foreground suggests the... | 1556659852054634496 | The video depicts a severe nighttime flood scene, likely caused by a heavy rainstorm, in an urban area. The floodwaters rush through the streets, creating turbulent waves and foam, and the area is illuminated by streetlights. The viewer observes the scene from an elevated or protected position, indicated by a railing in the foreground. Throughout most of the video, there are no visible people or vehicles, emphasizing the severity of the flood and the lack of immediate human activity.
As the video progresses, a large tree is shown partially submerged in the water, and later, bicycles are visible in the foreground, suggesting the location might be a residential area or a common gathering spot for locals. Eventually, a person holding an umbrella appears, standing on an elevated walkway and observing the rain, indicating that some individuals are still present in the area despite the challenging weather conditions.
The scene shifts to show the person walking along the elevated walkway, and later, buildings with illuminated windows are visible in the background, suggesting that some people are still inside. The presence of a vehicle and later two people holding umbrellas walking along the wet pavement highlights the ongoing activity and resilience of urban life amidst the storm.
Throughout the video, the turbulent water, streetlights, and illuminated buildings create a vivid depiction of the flood's impact on the urban environment. Despite the absence of an original ASR transcript, the visual elements convey a powerful narrative of a community coping with severe weather conditions. | {
"original": "",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Not Available",
"translated_llm": "- What?",
"translated_whisper": ""
} |
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, S... | [
"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 Distri... | [
"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 grou... | {
"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"
} | 32 | [
"The current frame shows a flooded staircase inside a building. Water is flowing down the steps, creating a cascading effect. The walls and handrails on either side of the staircase are visible, and the water has reached a significant level at the bottom, indicating heavy flooding.",
"The current frame depicts a ... | 1556995436052873218 | A severe flooding situation is depicted inside a building, where water is forcefully cascading down a staircase, creating a turbulent and chaotic scene. The flooding is intense and heavy, with high water levels persisting throughout the area. The water flow remains continuous and powerful, emphasizing the severity of the flooding situation. The scene is dominated by the sound of rushing water, but the audio transcript only contains a single, unclear utterance, "What?" suggesting that the video may be more focused on visually conveying the extent of the flooding rather than providing a narrative or dialogue. The overall context suggests a serious flooding event that is causing significant disruption and damage to the building. | {
"original": "",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "- What?",
"translated_llm": "- What?",
"translated_whisper": ""
} |
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, S... | [
"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 Distri... | [
"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 grou... | {
"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"
} | 32 | [
"The current frame shows an elevator with the number \"29\" displayed above the door. The text \"서초아크로비스타\" is visible at the bottom of the image, suggesting the name of the building or area. The elevator doors are closed, and the interior appears to be dark, with no visible movement or people inside. The surroundi... | 1556833205671456768 | The video depicts a stationary elevator at the "서초아크로비스타" building, primarily showing the elevator doors closed at the basement level "B2". The surrounding area is made of marble, with control panels to the right of the elevator and a yellow safety area on the floor. Throughout the video, there are no visible movements or people inside or outside the elevator, and the doors remain closed for the majority of the time, with a brief moment where the doors open and then close again. The elevator's floor indicator changes from "29" to "23" and eventually to "B2", where it remains for the rest of the video. The lack of movement or activity suggests the video might be a static observation or a waiting period. | {
"original": "",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Not Available",
"translated_llm": "- What?",
"translated_whisper": ""
} |
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, S... | [
"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 Distri... | [
"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 grou... | {
"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"
} | 32 | [
"The current frame shows people walking on a city street, with one person holding a black umbrella. The scene appears to be bustling, possibly during a typhoon season as suggested by the ASR transcript. The presence of umbrellas and wet pavement indicates rainy weather. The surroundings feature buildings and cars, ... | 1557013913216352258 | The video depicts a flooded urban street in Korea during the typhoon season. The scene shows people navigating through the flooded area, holding umbrellas and wading through the water, as they try to go about their daily activities despite the challenging conditions. Vehicles are also seen driving through the flooded street, creating ripples and splashes. The reflections of city lights on the water's surface create a shimmering effect, highlighting the contrast between the dark, wet surroundings and the ongoing city activity. The video captures the severity of the flooding, with water covering the road and causing difficulties for both pedestrians and vehicles. The ASR transcript emphasizes the unusual conditions, describing the streets as "blue" and encouraging the viewer to "take a look" at the situation, underscoring the significant impact of the typhoon on the city. Overall, the video provides a glimpse into the challenging conditions faced by city residents during the typhoon season in Korea. | {
"original": " Take a look at typhoon season in Korea The streets are blue Take a look at that",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Take a look at the typhoon season in Korea. The streets are blue. Take a look at that.",
"translated_llm": "Take a look at the typhoon season in Korea The streets are blue Take a look at that",
"translated_whisper": " Look at those typhoon-themed in Korea. The streets are blue. Look at that."
} |
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... | [
"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 A... | [
"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 respond... | {
"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"
} | 32 | [
"Emergency responders are present at the site of a large, damaged building, which appears to have suffered significant fire damage. A fire truck and several ambulances are positioned on the road nearby. People, likely officials or emergency personnel, are gathered around, and construction materials are visible in t... | SCkWqQ49yUQ | The video depicts the aftermath of a severe fire at the Icheon logistics center, where emergency responders and officials are actively engaged in disaster response efforts. The Icheon city government has established a disaster safety countermeasures headquarters, led by Um Tae-jun, comprising a 30-member team with 11 specialized teams, including medical relief and equipment support. The team is involved in patient transportation and on-site search operations, with a focus on rescuing those trapped inside the damaged building. The fire has resulted in significant damage to the logistics center, with 38 people reported dead and 8 missing. The video showcases the coordinated response efforts, highlighting the severity of the fire incident and the ongoing recovery operations. Throughout the video, the emphasis remains on the disaster response efforts, the establishment of the disaster safety countermeasures headquarters, and the active involvement of the 30-member team in managing the aftermath of the fire. | {
"original": " 이천 물류센터 화재와 관련해 관할 지자체가 재난안전대책본부를 구성했습니다. 경기 이천시는 엄태준 실장을 본부장으로 의료구호반과 장비지원반 등 11개반으로 꾸려진 30명 규모의 재난안전대책본부를 편성해 환자 이송과 현장 수색작업 등에 참여합니다.",
"refined": "Icheon city has set up a disaster safety countermeasures headquarters in response to the fire at the logistics center. The city has organized a 30-member disaster safety countermeasures headquarters, consisting of 11 teams including a medical relief team and equipment support team, with Um Tae-jun as the head, to participate in patient transportation and on-site search operations.",
"summarized": "A disaster response headquarters has been formed by the local government in response to a fire at the Icheon logistics center. The headquarters, led by Director Eom Tae-joon, consists of 11 teams, including a medical rescue team and an equipment support team, with a total of 30 members, and is participating in patient transportation and on-site search operations.",
"translated": "Regarding the fire at the Icheon logistics center, the local government has established a disaster safety countermeasures headquarters. The city of Icheon has organized a 30-member disaster safety countermeasures headquarters, consisting of 11 teams, including a medical relief team and an equipment support team, with Um Tae-jun as the head, to participate in patient transportation and on-site search operations.",
"translated_llm": "The city of Gievanshah participates in the 30 person emergency response center, which is composed of 11 units, including emergency medical and equipment support.",
"translated_whisper": " The local branch with its Rural Transportation Centre is the national emergency solution resource."
} |
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... | [
"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 A... | [
"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 respond... | {
"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"
} | 32 | [
"The current frame depicts a large building engulfed in thick, dark smoke, indicating a significant fire. There is visible water being sprayed onto the structure from the front, suggesting firefighting efforts are underway. The intense smoke and active fire response imply a serious situation, possibly involving eme... | oRy1PP3KzDs | A large fire broke out at a four-story logistics warehouse under construction in Icheon, Gyeonggi-do, on the 29th, resulting in significant damage and loss of life. The fire, which started in the basement, is suspected to have been caused by the evaporation of urethane foam during work. At the time of the fire, 78 people were working inside the building. Firefighters actively worked to control and extinguish the blaze, spraying water onto the structure from the front. Unfortunately, the incident resulted in multiple fatalities, with 11 confirmed deaths and 14 people initially reported as out of contact. The situation worsened, with the number of fatalities potentially increasing to over 25. Deceased individuals were discovered during the search process underground. The video, sponsored by the Gyeonggi-do Fire and Disaster Headquarters, appears to be a news report covering the incident, highlighting the severity and urgency of the situation as firefighters battled to extinguish the fire and address the crisis. | {
"original": " Hello everyone, welcome to my channel. Today I will show you how to make a delicious and delicious chicken breast with chicken breast . I will show you how to make a delicious chicken breast with chicken breast. First, I will make a chicken breast. I will make a chicken breast with chicken breast. I will make a chicken breast There is also a",
"refined": "Hello everyone, welcome to my channel. Today I will show you how to make a delicious chicken breast. First, I will cut the chicken breast into small pieces.",
"summarized": "The host welcomes viewers to their channel and introduces a recipe for making a delicious chicken breast. They plan to demonstrate the cooking process, starting with cutting the chicken breast into small pieces.",
"translated": "Hello everyone, welcome to my channel. Today I will show you how to make a delicious chicken breast. First, I will make a chicken breast.",
"translated_llm": "Hello everyone, welcome to my channel. Today I will show you how to make a delicious and delicious chicken breast with chicken breast. I will show you how to make a delicious chicken breast with chicken breast. First, I will make a chicken breast. I will make a chicken breast with chicken breast. I will make a chicken breast.",
"translated_whisper": " Hello everyone, welcome to my channel. Today I will show you how to make a delicious and delicious chicken breast with chicken breast . I will show you how to make a delicious chicken breast with chicken breast . First, cut the chicken breast into small pieces . To the right is the Hale's Bay, which used to exist. The entire beach, or the Hale's Bay, is just a small town. With all the"
} |
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... | [
"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 A... | [
"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 respond... | {
"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"
} | 32 | [
"A large building is engulfed in flames with thick black smoke billowing out of it. Firefighters are actively working to extinguish the fire, using hoses to spray water onto the structure. The intense smoke indicates a significant fire, and the presence of firefighting equipment suggests an ongoing effort to contai... | rqRqdg-oxkU | A large fire broke out in a multi-story building, with thick black smoke billowing out and visible flames. Firefighters responded quickly, using hoses to spray water onto the structure in an effort to control and extinguish the blaze. The fire was intense, causing significant damage to the building, and the firefighters worked diligently to contain it. Despite their efforts, the fire remained active for an extended period, with smoke and flames visible throughout. The atmosphere was tense as the firefighters battled the blaze, utilizing various equipment to bring it under control. The exact cause of the fire and the extent of the damage are unclear, but the situation was critical, requiring a substantial response from the firefighting team. | {
"original": " Day 2 They want to kill me Terima kasih telah menonton",
"refined": "They want to kill me.",
"summarized": "The speaker, a K-pop star, is introducing a story that spans multiple generations of K-pop, starting from the first generation and moving on to the second, which was introduced in a previous part (Day 1) and is continuing into the current part (Day 2).",
"translated": "Day 2: They want to kill me. Thank you for watching.",
"translated_llm": "Day 2 They want to kill me Thank you for watching",
"translated_whisper": " Hello, this is the K-pop star from the 3rd generation of K-pop. Let's start from the second generation that was just introduced. This is a story that was published in 2015. So I'll start from the first generation and go to the second generation. Hey, this is for the camera. So who'll film the beginning of the second generation?"
} |
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... | [
"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 A... | [
"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 respond... | {
"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"
} | 32 | [
"The image shows a large plume of black smoke rising from a building at a water treatment facility in Icheon. The smoke suggests a fire or explosion has occurred at the site. The Korean text in the image translates to \"Currently at the Icheon Water Treatment Facility.\" The scene depicts emergency response activit... | 6Z2tI74NGGQ | A severe fire broke out at the Icheon water treatment facility, causing significant damage and loss of life. The incident began with an explosion, and the fire quickly spread, engulfing a large building and emitting thick black smoke. Emergency responders, including firefighters and police officers, were quickly deployed to the scene to contain the blaze and search for survivors. Despite their efforts, the fire proved difficult to control, and the situation was exacerbated by the risk of further explosions. The facility suffered substantial structural damage, with walls and windows destroyed, and debris scattered throughout the area. The emergency response efforts continued, with firefighters working to extinguish the flames and investigate the cause of the explosion. The incident resulted in casualties, and the aftermath was marked by a sense of chaos and tragedy, with the facility's completion delayed due to the devastating event. | {
"original": " . 요즘 폭발하는 모습을 시정하고 있습니다.",
"refined": "Recently, they are dealing with the aftermath of the explosion.",
"summarized": "A recent incident involving an explosion is being addressed, with the focus on helping victims who were injured, including a man burned by a broken window, and taking steps to prevent future damage.",
"translated": "I'm currently fixing my explosive behavior.",
"translated_llm": "I'm fixing the exploding look these days.",
"translated_whisper": " At the site where the water was flowing, They started to see thick smoke as soon as going to head. It's all because of the smoke. In fact, it was already the first time to see firework, when there are only few individuals. In the end, there is no problem. The damage happened to people have all the clothes burned So now he is fixing it To save the lives of the victims To open up a new world A man who got burned by a broken window Is going back to a new home"
} |
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... | [
"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 A... | [
"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 respond... | {
"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"
} | 32 | [
"The image shows a large plume of black smoke rising from an area with buildings and construction sites, suggesting a fire or explosion has occurred. The surrounding area appears to be an industrial or construction site. The visible text in the image is in Korean and reads \"이천 물류센터 화재 현장\" which translates to \"Ic... | JCKK0gu12zI | A large fire broke out at the Icheon Logistics Center in South Korea, with thick black smoke billowing from the building. Firefighters actively worked to control the blaze, directing strong water streams from hoses towards the fire. The severity of the fire was evident from the extensive smoke and the ongoing efforts of the emergency response team. The fire was one of two significant incidents to affect the country in the past three days, as mentioned in the audio. The video depicted the firefighting efforts, with firefighters persisting in their attempts to extinguish the fire amidst challenging conditions. The scene was filled with industrial materials and equipment, highlighting the logistical nature of the facility. The text "이천 물류센터 화재 현장" consistently confirmed the location and nature of the event as the fire scene at the Icheon Logistics Center. | {
"original": " The biggest earthquake in China has hit the country and has hit the world. The earthquake hit the world twice in the last three days and three days Terima kasih telah menonton",
"refined": "Not Available",
"summarized": "Not Available",
"translated": "The biggest earthquake in China has hit the country and has affected the world, striking the world twice in the last three days, thank you for watching.",
"translated_llm": "The biggest earthquake in China has hit the country and has hit the world.",
"translated_whisper": " In Korea today, many Chinese people like to go to the market for a good price. Some of these Chinese people are a bit poor, but there are also some who choose to stay in the market and go to the market in groups. It is an old saying that I can only go to the market in groups that I really like. The city of San Francisco is home to the most populous and most populous cities in the world. The city is home to the largest populous city in the world."
} |
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... | [
"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 A... | [
"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 respond... | {
"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"
} | 32 | [
"A news anchor is standing in front of a backdrop featuring a map and text that appears to be in Korean. The anchor is delivering a news report, likely discussing the interim investigation results of a fire at a logistics warehouse in Icheon, Gyeonggi Province, which resulted in 38 deaths and 10 injuries. The polic... | ZFWsFwpfGlE | A news report announces the interim investigation results of a devastating fire at a logistics warehouse in Icheon, Gyeonggi Province, which resulted in 38 deaths and 10 injuries. The fire, presumed to have been caused by an explosion triggered by an unidentified fire during urethane work on the second basement level, led to a thorough investigation by the police. The investigation involved four large-scale on-site inspections and the questioning of around 80 individuals related to the fire. As a result, 17 individuals were charged with occupational negligence resulting in death. The video depicts the aftermath of the fire, with the severely damaged warehouse and surrounding area filled with debris and charred materials. Investigators in protective suits are shown examining the site, sifting through rubble, and collecting evidence. The scene emphasizes the severity of the incident and the ongoing efforts to understand the cause of the fire and to hold those responsible accountable. Throughout the video, the focus remains on the tragic outcome and the detailed process of investigation, underscoring the serious nature of the incident and the commitment to justice for the victims. | {
"original": " 경기 남부지방경찰청이 오늘 오전 38명의 목숨을 앗아간 경기도 이천 물류창고 화재 사고에 대한 중간 수사 결과를 발표합니다. 지난 4월 29일 경기 이천시의 한 물류창고 공사 현장에서 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.",
"refined": "The South Gyeonggi Provincial Police Agency announced the mid-term investigation results of the fire at a logistics warehouse in Icheon, Gyeonggi Province, which claimed 38 lives. On April 29, a fire broke out at a logistics warehouse construction site in Icheon, killing 38 people and injuring 10. The cause of the fire is believed to be an explosion that occurred during urethane work on the second basement floor, resulting from an unknown ignition source. The police conducted four large-scale on-site inspections and investigated around 80 people related to the fire, applying charges of job-related negligent homicide to 17 of them.",
"summarized": "The South Gyeonggi Province Police Department announced the mid-term investigation results of the Icheon logistics warehouse fire that killed 38 people and injured 10. The fire is believed to have been caused by an explosion during underground urethane work. Following four large-scale field inspections and the investigation of around 80 people related to the fire, the police have applied charges, including job-related negligent homicide, to 17 individuals.",
"translated": "The Gyeonggi Southern Provincial Police Agency announced the interim investigation results of the fire at a logistics warehouse in Icheon, Gyeonggi Province, which claimed 38 lives. On April 29, a fire broke out at a logistics warehouse construction site in Icheon, resulting in 38 fatalities and 10 injuries. The cause of the fire is presumed to be an explosion triggered by an unidentified fire that occurred during urethane work on the second basement level. The police conducted four large-scale on-site inspections and investigated around 80 individuals related to the fire, applying charges of occupational negligence resulting in death to 17 of them.",
"translated_llm": "The Southern District Police Department announced the results of the interim investigation into the fire incident in the Gujarat City on 29 April, which claimed 38 lives. The fire at a warehouse in Gujarat city on 29 April caused 38 deaths and 10 injuries. The cause of the fire is believed to have been caused by an explosion in a fireplace during underground 2 storey construction.",
"translated_whisper": " The South Gyeonggi Province Police Department announced the mid-term investigation results of the fire in Gyeonggi-do's Icheon logistics warehouse that killed 38 people this morning. On April 29, 38 people were killed at a logistics warehouse construction site in Icheon, Gyeonggi Province, and 10 people were injured. The cause of the fire is expected to be an explosion caused by a fire in a low-rise garden that occurred during the underground urethane work. The police investigated 80 people involved in the fire, including four large-scale field inspections, and applied for a job-related, excessive death penalty of 17 people."
} |
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... | [
"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 A... | [
"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 respond... | {
"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"
} | 32 | [
"A text animation displays \"Right now\" on a dark background, indicating the start of a video segment.",
"The video shows an aerial view of a large building on fire. Dark smoke billows into the sky, and the surrounding area appears to be under construction. The text in the image reads \"4월 29일 경기도 이천 물류창고 화재 현장,... | 2SCMvAXvBLM | A large industrial building in Icheon, Gyeonggi-do, is engulfed in flames, with thick, dark smoke billowing into the sky. The surrounding area, which appears to be under construction, is filled with emergency service vehicles and personnel actively working to combat the fire. The scene is captured from an aerial view, showing the severity of the fire and the ongoing emergency response efforts. The video is presented as a live or breaking news update, indicated by the text "Right—NOW" visible in the top right corner of several frames. The fire continues to burn intensely throughout the video, with emergency services present and working to address the situation. The video concludes with a transition segment, displaying a circular logo with the text "Right—NOW" on a two-tone background, signaling the end of the current content and potentially leading into the next video. | {
"original": " 다음 영상에서 만나요.",
"refined": "See you in the next video.",
"summarized": "The speaker bids farewell and thanks the viewer for watching.",
"translated": "See you in the next video.",
"translated_llm": "See you in the next video.",
"translated_whisper": " Thank you for watching."
} |
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... | [
"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 A... | [
"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 respond... | {
"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"
} | 32 | [
"In the current frame, two news anchors are presenting a report in a studio. Behind them is an image of a fire at a logistics warehouse construction site in Icheon, Gyeonggi Province. Emergency services are visible at the scene, indicating ongoing response efforts. The news is discussing the police's thorough searc... | n8dcNr14bm8 | The video depicts a news report and investigation of a devastating fire at a logistics warehouse construction site in Icheon, Gyeonggi Province, where 38 people lost their lives. The Southern Gyeonggi Provincial Police Agency's Scientific Investigation Unit is conducting a thorough and meticulous search of the site, utilizing 15 investigators and heavy equipment to locate the unrecovered remains and personal belongings of the victims. The search efforts are ongoing, with investigators in protective suits carefully examining the debris-littered site, and heavy equipment being used to assist in the operation. The police are working diligently to ensure a comprehensive search of the site, with a focus on finding the remaining victims and their belongings. The video shows the extensive damage to the building, with charred and blackened walls, and debris scattered throughout the area. The investigation is a precise and thorough operation, with the police taking a meticulous approach to searching the site and recovering the remains and personal belongings of the victims. | {
"original": " 38명이 숨진 경기도 이천 물류창고 공사장 화재 현장에서 경찰이 아직 수습되지 않은 유해 일부와 유류품 등을 찾기 위한 정밀 수색을 하고 있습니다. 경기 남부지방경찰청 과학수사대는 오늘 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. 감사합니다.",
"refined": "The police are conducting a thorough search at the Icheon logistics warehouse construction site, where 38 people died in a fire, to find the uncollected remains and personal belongings. The Gyeonggi Nambu Provincial Police Agency's scientific investigation team has deployed heavy machinery and 15 forensic investigators since 9 am today to search the site in Icheon's Moga-myeon. Today's search is focused on finding the remaining uncollected remains and personal belongings of the victims who died in the accident.",
"summarized": "The police are conducting a thorough search at the Icheon logistics warehouse construction site, where 38 people died in a fire, to find the remaining unsecured remains and personal belongings of the victims, with 15 science investigators and heavy equipment involved in the search.",
"translated": "At the scene of the fire at a logistics warehouse construction site in Icheon, Gyeonggi Province, where 38 people lost their lives, the police are conducting a thorough search to find the remaining unrecovered remains and personal belongings. The Southern Gyeonggi Provincial Police Agency's Scientific Investigation Unit has deployed 15 investigators and heavy equipment since 9 am today to search the construction site in Moga-myeon, Icheon. The focus of today's search is to find the unrecovered remains of the victims who died in the accident and their personal belongings.",
"translated_llm": "38 people are being searched at the scene of the incident. Police are searching for untrained personnel and supplies. The Southern District Police Science and Research Unit has deployed 15 people to search the scene of the incident.",
"translated_whisper": " 38 people are being searched for illegal items and 15 people including the center equipment and science investigators are being put in and the site of the construction of the 2,000-seat logistics warehouse is being thoroughly searched. Today's search is a method that focuses on finding the unsecured part and the remains of the victims who died 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... | [
"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 A... | [
"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 respond... | {
"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"
} | 32 | [
"The image depicts a large building, likely a warehouse, with emergency vehicles, including ambulances and a fire truck, parked in front. The scene indicates an ongoing investigation, as the text mentions that a task force has been established to conduct a joint on-site inspection with the National Forensic Science... | Z1TmlndGWrg | A fire incident occurred at the Icheon mud resin factory, prompting a large-scale emergency response and investigation. The Southern Gyeonggi Provincial Police Agency established a 125-member task force, including metropolitan and scientific investigation units, to investigate the incident. A joint on-site inspection with the National Forensic Science Institute and fire authorities was scheduled for 10:30 AM to gather evidence and determine the cause of the fire. Emergency responders, including firefighters, police officers, and medical professionals, were deployed to the scene to provide medical attention and support the investigation. The investigation unit worked to coordinate efforts and conduct a thorough examination of the fire-damaged building, with a focus on determining the cause of the incident and ensuring public safety. The response involved multiple agencies and units, highlighting the complexity and severity of the incident. | {
"original": " 경찰은 전담 수사본부를 편성했습니다. 경기 남부지방 경찰청은 광역수사대와 과학수사대 등 125명 규모의 전담 수사본부를 편성했다고 했습니다. 수사본부는 오전에 오늘 오전 10시반에 국립과학수사연구원, 또 소방당국과 함께 합동으로 현장 감식을 진행할 예정입니다.",
"refined": "The police have formed a dedicated investigation department. The Gyeonggi Southern Provincial Police Agency has established a 125-member investigation department, including a large-scale investigation unit and a scientific investigation unit. The investigation department plans to conduct a joint on-site inspection with the National Forensic Service and the fire authorities at 10:30 this morning.",
"summarized": "The police have formed a dedicated investigation unit, comprising 125 members including the scientific investigation unit and others, to handle the case. A joint on-site inspection is scheduled to take place at 10:30 this morning, in collaboration with the National Science Investigation Institute and the fire department.",
"translated": "The police have established a dedicated investigation unit. The Southern Gyeonggi Provincial Police Agency has formed a 125-member task force, including a metropolitan investigation unit and a scientific investigation unit. The investigation unit plans to conduct a joint on-site inspection with the National Forensic Science Institute and the fire authorities at 10:30 this morning.",
"translated_llm": "The police have set up a 125-person task force of the Southern District Police Department, including the Canton Water Department and the Science Water Department.",
"translated_whisper": " Police formed The National Investigation Department The North Korean Police Department of Gyeonggi Province formed the investigation group of 125-person such as the Defense Investigation Unit of a department on nuclear energy At 10 o'clock this morning, the investigation department is scheduled to run a field inspection with the National Science Investigation Institute and the fire department."
} |
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... | [
"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 A... | [
"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 respond... | {
"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"
} | 32 | [
"The image shows an aerial view of a large building engulfed in thick, black smoke, indicating a severe fire. Firefighters and emergency vehicles are present, actively working to control and extinguish the blaze. The building is located near a road with several cars parked nearby, and the surrounding area is partia... | tHSh4jwdIJg | A severe fire broke out at an industrial facility during urethane foam work, spreading rapidly and engulfing the building in thick, black smoke. Emergency responders, including firefighters, were quickly deployed to the scene, working to control and extinguish the blaze. The fire caused significant damage to the building and posed a risk to nearby vehicles and personnel. Despite challenges, including dense smoke underground that made it difficult to approach the area, emergency responders managed to contain the fire. The incident resulted in 25 workers being injured. The response efforts were extensively documented, with media personnel present to capture the event. The fire was eventually brought under control, but not before causing substantial damage and disruption to the industrial site, affecting the normal work of 800 people. The video documentation of the incident includes footage of the fire, emergency response efforts, and the aftermath, with the source of the video attributed to "비디오머그" (Video Mug). | {
"original": " 자막러 연소 확산이 굉장히 빠른 것으로 이루어진 것으로 봐서 오레탄 폼 작업 중에 화재가 발생한 것으로 추정하고 있습니다.",
"refined": "The fire is believed to have occurred during the urethane foam work as the fire spread was very fast.",
"summarized": "A fire occurred at a location, likely during long-term foam work, as evidenced by the rapid spread of the fire.",
"translated": "The fire is believed to have occurred during the urethane foam work, given that the subtitles show the fire spreading very quickly.",
"translated_llm": "The fire is believed to have started during the oil shale process.",
"translated_whisper": " An old cafe is located in Pohang There are many new cafes and restaurants in the old town The cafe is small but there are so many people I felt like buying ice cream There is a place right next to the cafe Today we are going on a trip to Pohang There are lots of delicious food Is this a joke? What kind of food do they have? The fire is expected to be a fire that occurred during the long-term foam work as the fire spread was very fast. The fire is expected to be a fire that occurred during the long-term foam work."
} |
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 d... | [
"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 ... | [
"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... | {
"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"
} | 32 | [
"In the current frame, a news reporter stands in front of a screen displaying footage of a damaged building. The screen shows the exterior of a building with broken windows and a ladder leaning against the structure. The text on the screen translates to \"A room turned into 'charred trash'... Daegu law office in di... | I2CxZCzE5pE | A severe fire broke out at a law office in Daegu, causing extensive damage to the second floor where the blaze originated. The fire left the area charred beyond recognition, with the 203rd office being particularly affected, its original shape unrecognizable due to the severe burning. Everything on the second floor, including the floor, stairs, and every step, was blackened, with windows shattered and black ash and glass shards scattered everywhere. In contrast, the first floor remained relatively intact. The fire caused significant structural damage, with ceiling material collapsing and wires hanging down, making it impossible to see without a flashlight during the daytime. Office staff members who narrowly escaped the fire returned to retrieve documents related to ongoing lawsuits, facing the challenge of salvaging what they could from the devastated offices. The incident has left those who lost their livelihoods in the fire wondering when they can return to their normal lives, with a sense of unease and uncertainty prevailing among the survivors. The report concludes with an emphasis on the severity of the fire's impact and the ongoing efforts of those affected to recover and move forward. | {
"original": " 지금부터는 어제 대구의 한 변호사 사무실에서 난 화재에 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. 먼저 이승환 기자입니다. 바닥부터 계단 한 칸 한 칸, 검게 그을리지 않은 곳이 없습니다. 유리창은 완전히 깨졌고, 곳곳에 검은 재와 유리 조각이 뒤섞인 채 쌓여 있습니다. 비교적 멀쩡한 1층과 달리 2층은 상태가 심각합니다. 불이 시작된 203호는 형체를 알아볼 수 없을 정도로 검게 타버렸습니다. 천장재가 떨어져 내리면서 전선이 아래로 늘어졌습니다. 대낮인데도 손전등 없인 한치 앞도 보이지 않습니다. 사무실과 사무실을 구분하는 가벽도 뚫려버렸습니다. 급히 대피하느라 몸만 빠져나온 다른 사무실 직원들도 오늘 다시 현장을 찾았습니다. 소송 관련 서류 등을 챙기기 위해서입니다. 각종 집기를 닦은 흰 천은 새카매졌습니다. 당시 건물 안에 있다 탈출한 변호사는 지금도 아찔하기만 합니다. 순식간에 삶의 터전을 잃어버린 사람들. 언제쯤 평범한 일상으로 돌아갈 수 있을지 막막하기만 합니다. JTBC 이승환입니다.",
"refined": "From now on, we will report on the latest developments in the fire that broke out at a law firm in Daegu yesterday. Our reporters obtained footage from inside the building where the accident occurred. The second floor, where the fire started, was severely damaged, with everything from the floor to the stairs charred beyond recognition. Reporter Lee Seung-hwan has more. The entire floor is blackened, with no area left untouched. The windows are completely shattered, and broken glass and soot are scattered everywhere. In contrast to the relatively intact first floor, the second floor is in a dire state. The 203rd office, where the fire originated, is unrecognizable due to the extensive fire damage. The ceiling material has collapsed, causing wires to dangle down. Even in daylight, visibility is near zero without a flashlight. The partition walls between offices have also been destroyed. Other office workers who narrowly escaped the fire returned to the scene today to collect documents related to ongoing lawsuits. The white cloth used to wipe various objects is now stained with soot. The lawyer who escaped from the building at the time still feels shaken. The fire has left people's lives in disarray, and it remains uncertain when they will be able to resume their normal lives. This is Lee Seung-hwan reporting for JTBC.",
"summarized": "A fire broke out at a lawyer's office in Daegu, South Korea. The second floor of the building was severely damaged, with walls and furniture blackened by the fire. The office where the fire started, room 203, was particularly badly damaged, with ceiling material fallen and wires exposed. The building's occupants, including lawyers and staff, escaped in a hurry, but were left shaken. They returned to the scene to collect important documents and belongings. The incident has left many people wondering when they can return to their normal lives.",
"translated": "From now on, we will report on the latest developments regarding the fire that broke out at a law office in Daegu yesterday. Our reporting team has obtained footage of the interior of the building where the accident occurred. The second floor, where the fire started, was severely damaged, with everything charred beyond recognition. Reporter Lee Seung-hwan has more. The floor, stairs, and every step were blackened, with no area left unscathed. The windows were completely shattered, and black ash and glass shards were scattered everywhere. In contrast to the relatively intact first floor, the second floor was in a dire state. The 203rd office, where the fire originated, was burned so severely that its original shape was unrecognizable. The ceiling material had collapsed, causing the wires to hang down, and even in the daytime, it was impossible to see without a flashlight. The partition walls between offices had also been breached. Other office staff members, who had narrowly escaped, returned to the scene today to retrieve documents related to ongoing lawsuits. A white cloth used to wipe down various equipment was now stained black. A lawyer who had managed to escape from the building at the time of the fire still can't shake off the feeling of unease. People who lost their livelihoods in an instant are now left wondering when they can return to their normal lives. This is Lee Seung-hwan reporting for JTBC.",
"translated_llm": "From now on, I will share with you what we have newly covered in relation to the fire in the lawyer's office in the big city yesterday. Our reporter has received video of the inside of the building that caused the accident. The second floor of the building that started the fire was so badly burned that it was difficult to identify the body. Firstly, the fire was a passing news reporter. From the floor, there was a staircase, a section, and nowhere not marked with a sword. The glass window was completely broken, and everywhere was piled with black ash and glass fragments mixed together.",
"translated_whisper": " From now on, we will tell you about the new news that we recently reported about a fire in a lawyer's office in Daegu yesterday. Our reporters took a video of the building where the accident happened. The second floor of the building where the fire started was so black that it was difficult to recognize the shape. First, reporter Lee Seung-hwan. There is no place that has not been blackened from the floor to the stairs. The glass window is completely broken and blackened and glass pieces are piled up in various places. Unlike the relatively normal first floor, the second floor is in a serious condition. The 203rd floor, where the fire started, burned black enough to make it impossible to recognize the shape. As the ceiling material fell, the wires were stretched down. Even though it's very low, there is no one-foot pressure without a flashlight. The wall between the office and the office was also pierced. Other office workers who were in a hurry to escape were also back on site today. to collect the files related to the lawsuit. The white cloth that wiped off various house keys has become black. The floor, records, and house keys are all black. The job itself is impossible. The lawyer who escaped from the building at the time is still dizzy. It's a little scary. Honestly, we were here for a few seconds, and there were about 10 people on the 5th floor, and I think they all died. I think we should be more careful in the future. People who lost their way in life in an instant. I'm just wondering when we'll be able to go back to normal life. This is Lee Seung-hwan of JTBC."
} |
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 d... | [
"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 ... | [
"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... | {
"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"
} | 32 | [
"The current frame shows two news anchors, a man on the left and a woman on the right, standing in a studio with a world map background. The man is wearing glasses, a dark suit, and a striped tie, while the woman is dressed in a maroon blazer. The screen displays the headline about a fire at a lawyer's office near ... | oVzTD-Qy5HM | A devastating fire broke out in a lawyer's office near the Daegu District Court, resulting in the deaths of seven people, including one suspected arsonist, and injuring over 40 others. The fire started on the second floor of the building, which has two basement levels and five above-ground levels, at around 10:55 am. Despite being extinguished within 20 minutes, the fire spread rapidly due to the lack of a sprinkler system on the above-ground floors, leading to significant casualties. All victims were found in room 203 on the second floor. The police are investigating the incident, reviewing CCTV footage of a 50-year-old suspect who may have set the fire, and have formed a dedicated team to determine the cause. The fire department's urgent actions, including breaking windows to rescue those trapped, underscore the severity of the situation. The incident occurred in an area densely populated with lawyer's offices, and the building's limited fire safety measures are under scrutiny. Emergency responders and investigators are working together to understand the circumstances surrounding the fire and to provide support to those affected. | {
"original": " 오늘 낮 대구 지법 근처의 한 변호사 사무실에서 방화로 보이는 화재가 발생해 방화용지자 1명을 포함해 7명이 숨지고 40여 명이 부상을 입었습니다. 자세한 소식, 현장에 취재기자 연결해서 들어보겠습니다. 정지훈 기자. 대구 변호사 사무실 빌딩 화재 현장에 나와 있습니다. 제 바로 뒤편 건물을 보시면 건물 전면 유리창들이 깨진 걸 보실 수 있을 텐데요. 신고를 받고 출동한 소방당국이 화재 진압과 구조를 위해서 일부러 깨트린 겁니다. 당시 상황이 얼마나 긴박했는지를 알 수 있는데요. 이 건물은 지하 2층과 지상 5층짜리 건물 중 오늘 오전 10시 55분쯤 건물 2층에서 불이 시작됐습니다. 불은 소방당국에 의해 20여 분 만에 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 들렸다는 신고가 접수됐습니다. 사망자의 신호는 아직 확인되지 않았습니다. 같은 장소에서 방아 용의자도 숨진 채 발견됐습니다. 불이 나자 건물 3층에 있던 사람들은 4층 등 건물 위층으로 일부는 건물 밖으로 설치된 비상계단을 통해 옥상으로 대피하기도 했습니다. 화재 현장에 있었던 목격자의 말 들어보시겠습니다. 방 터지면서 유리창 깨지는 소리 계속 나고 제가 너무 놀라가지고 상담실에서 뛰쳐나가가지고 문 딱 열라고 하니까 문이 이미 뜨거웠어요. 이미 뜨거워서 제가 문을 몸으로 밀쳐가지고 문을 열고. 소방당국은 화재 진압 과정에서 급격히 연소가 확대돼 인명피해 규모가 커졌다고 설명했습니다. 당국은 브리핑에서 화재 건물 지하층에만 스프링클러가 있고 지상층에는 스프링클러가 설치돼 있지 않다고 밝혔습니다. 화재가 발생한 빌딩은 법원 뒤쪽에 위치해 변호사 사무실이 밀집한 곳입니다. 경찰은 50대 방아 용의자가 CCTV상 주거지에서 뭔가를 들고 나오는 장면을 확인하고 자세한 화재 경위를 조사하고 있는데요. 경찰은 불이 난 변호사 사무실에서 의뢰된 송사 사건의 상대인 용의자가 사무실에 불을 낸 것으로 보고 수사 전담팀을 편성해 사건 발생 경위 등을 집중 조사할 방침입니다. 지금까지 태구에서 연합뉴스TV 청준입니다.",
"refined": "A fire broke out at a lawyer's office near the Daegu courthouse today, resulting in the deaths of seven people, including one arson suspect, and injuring over 40. For more details, let's connect with our reporter on the scene, Jung Ji-hoon. I'm reporting live from the scene of the fire at the Daegu lawyer's office building. As you can see behind me, the front windows of the building are shattered. The fire department intentionally broke them to put out the fire and rescue those inside, giving you an idea of how urgent the situation was. The building has two basement levels and five above-ground levels. The fire started on the second floor around 10:55 am today and was extinguished by the fire department within 20 minutes. However, seven people, including five men and two women, lost their lives, and over 40 suffered burns or smoke inhalation. The victims were all found in room 203 on the second floor. At the time of the fire, the 119 emergency services received reports of black smoke and explosion sounds coming from the second floor. The identities of the victims have not been confirmed yet. The arson suspect was also found dead at the scene. When the fire broke out, people on the third floor escaped to the upper floors, including the rooftop, using the emergency stairs installed outside the building. Let's hear from a witness who was at the scene of the fire. The fire department explained that the rapid spread of the fire during the extinguishing process led to a larger number of casualties. According to the authorities, the building's basement floor is equipped with a sprinkler system, but not the above-ground floors. The building where the fire occurred is located behind the courthouse in a densely populated area of lawyer's offices. The police are investigating the cause of the fire, reviewing CCTV footage of the suspect leaving their residence with an object. The police are also looking into the possibility that the suspect, who was involved in a case being handled by the lawyer's office, started the fire and have formed a task force to investigate the circumstances surrounding the incident.",
"summarized": "A fire broke out on the 2nd floor of a 7-story lawyer's office building near Daegu Court, killing 7 people, including 5 men and 2 women, and injuring over 40 others. The fire started at 10:55 am and was extinguished in 20 minutes. The victims were found on the 2nd floor, and a suspect, believed to be a 50-year-old man involved in a lawsuit, was also found dead at the scene. Witnesses reported hearing an explosion and seeing black smoke, and some people evacuated to the rooftop via emergency stairs. The fire department attributed the rapid spread of the fire to the lack of sprinklers on the upper floors. The police are investigating the incident, suspecting arson, and have formed a special task force to investigate the cause of the fire and the suspect's involvement.",
"translated": "A fire broke out in a lawyer's office near the Daegu District Court today, resulting in the deaths of seven people, including one suspected arsonist, and injuring over 40 others. Let's hear more about the incident from our reporter at the scene. Reporter Jung Ji-hoon is live at the Daegu lawyer's office building, where the fire occurred. Behind me, you can see that the front windows of the building are shattered. The fire department intentionally broke the windows to put out the fire and rescue those trapped inside, giving you an idea of how urgent the situation was. The building has two basement levels and five above-ground levels. The fire started on the second floor at around 10:55 am today. Although the fire was extinguished within 20 minutes by the fire department, seven people, including five men and two women, lost their lives, and over 40 others are being treated for injuries from the fire or smoke inhalation. The victims were all found in room 203 on the second floor. At the time of the fire, a report was made to the emergency services about black smoke and an explosion on the second floor. The identities of the deceased have not been confirmed yet. The suspected arsonist was also found dead at the scene. People who were on the third floor of the building evacuated to the upper floors, including the rooftop, using the emergency stairs installed outside. Let's hear from a witness who was at the scene of the fire. The fire department explained that the fire spread rapidly, resulting in a significant number of casualties. According to the authorities, the building where the fire occurred, which is located behind the court in an area densely populated with lawyer's offices, had a sprinkler system only on the basement level, but not on the above-ground floors. The police are investigating the incident, including a CCTV footage showing a 50-year-old suspect carrying something out of their residence. The police believe the suspect may have set the fire at the lawyer's office, which was handling a case against them, and have formed a dedicated team to investigate the cause of the fire. That's all for now from Daegu.",
"translated_llm": "The fire was reported to have broken out in a fire in a fire station near the city of Bruges, which was reported to have been a fire. The fire broke out in the basement of the building at around 10:55 am. The fire was reported to have broken out in the basement of the building at around 10:55 am. The fire was reported to have started in the basement of the building at around 10:55 am. The fire was reported to have started in the basement of the building at around 20 minutes after the fire was reported by the police, but the fire started in the basement of the building at about 10:55 am. The fire started in the basement of the building at around 10 minutes after the fire started.",
"translated_whisper": " A fire in a lawyer's office near Daegu today. Including one fireman, seven people were killed and more than 40 people were injured. Let's listen to the details in the newsroom. Reporter Jung Ji-yoon. Daegu Lawyer's Office Building Fire at the scene. If you look at the building right behind me, you can see that the front windows of the building are broken. The fire broke out on purpose to rescue the fire department. We can see how urgent the situation was at the time. This building is on the 2nd basement and 5th floor. The fire started on the 2nd floor at 10.55 a.m. today. The fire was put out in 20 minutes by the fire department. 7 people, including 5 men and 2 women, are breathing. More than 40 people are being treated by fire or smoke. The victims were found on the 2nd floor of 203. At the time of the fire, the building on the 2nd floor was covered with black smoke and the explosion was heard. The death signal has not been confirmed yet. Defendants were also found in the same place. The people on the 3rd floor of the building were on the 4th floor and the upper floor of the building. Some of them were evacuated to the rooftop through emergency stairs installed outside the building. Let's listen to the witness at the scene of the fire. The fire department said that the number of casualties has increased rapidly as the fire broke out. The party said that there is a sprinkler only on the basement floor of the fire building in the briefing, and that there is no sprinkler on the ground floor. The building where the fire occurred is located behind the court and is a close-knit place for the lawyer's office. The police are investigating the scene of a 50-year-old fireman checking something out of the parking lot above the CCTV and investigating the fire. The police are investigating the case of the suspect in the case of the investigation, which was requested by the police in the lawyer's office. This is Jung-joon from Daegu."
} |
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 d... | [
"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 ... | [
"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... | {
"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"
} | 32 | [
"A reporter stands outside near the Daegu District Court, reporting on a fire incident that occurred on the second floor of a building housing law offices. The fire resulted in seven fatalities and around 40 injuries. The scene shows emergency response personnel, including a fire truck, indicating ongoing efforts t... | 1534839444830044161 | A devastating fire broke out on the second floor of a densely populated building housing law offices near the Daegu District Court in Daegu, resulting in seven fatalities and around 40 injuries. The fire, which occurred at around 10:55 am, was met with a rapid response from the fire department, who deployed over 60 pieces of equipment and more than 160 personnel to extinguish the flames within 20 minutes. Witnesses reported an explosion and a large amount of smoke, which spread quickly due to the lack of essential fire safety measures such as sprinklers and fire shutters. Emergency responders, including firefighters and rescuers, worked diligently to rescue those trapped and manage the aftermath, with ongoing investigations into the cause of the fire, suspected to be arson due to dissatisfaction with a verdict. The incident highlights the urgency and severity of the situation, emphasizing the need for improved safety measures to prevent such tragedies in the future. | {
"original": " 일부가 깨져 있는데요. 신고를 받고 출동한 소방대원들이 화재 진압과 구조를 위해 깨트린 것으로 당시 긴박한 상황을 알 수 있습니다. 사고가 난 것은 오늘 오전 10시 55분쯤입니다. 대구 수성구 대구지방법원 인근의 한 변호사 사무실이 밀집한 건물 2층에서 불이 났습니다. 소방당국은 현재까지 인명피해가 사망 7명, 부상자 40여 명이라고 밝혔습니다. 주변 목격자들은 폭발음과 함께 많은 연기가 건물 외부로 뿜어져 나왔다고 진술했습니다. 소방당국은 장비 60여 대와 인력 160여 명을 투입해 불이 난 지 20여 분 만에 진화했습니다. 하지만 건물 지하층에만 스프링클러가 설치된 데다 방화 셔터가 없어 연기가 건물 내부로 순식간에 확산되면서 인명피해가 커진 것으로 보입니다.",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "A fire broke out on the second floor of a densely populated building that houses several law offices near the Daegu District Court in Suseong-gu, Daegu, at around 10:55 am today. The fire department, which received a report and responded to the scene, broke some parts of the building while putting out the fire and rescuing those trapped, reflecting the urgent situation at the time. According to the fire authorities, as of now, seven people have died and around 40 have been injured. Witnesses reported hearing an explosion and seeing a large amount of smoke billowing out of the building. The fire department deployed over 60 pieces of equipment and more than 160 personnel, managing to extinguish the fire within 20 minutes of its outbreak. However, it appears that the lack of sprinklers outside of the basement and the absence of fire shutters allowed the smoke to spread rapidly throughout the building, resulting in a greater loss of life.",
"translated_llm": "The incident occurred at around 10:55 am today. A lawyer's office in the vicinity of the Big City Water District Court building caught fire on the 2nd floor of a crowded building. So far, seven people have died and more than 40 injured.",
"translated_whisper": " A part of the building is broken. Firemen who were reported and dispatched were broken for fire and rescue. We can see the emergency situation. The accident happened around 10.55 a.m. A lawyer's office near Daegu's Suseong-gu District Court was set on fire on the second floor of the building. The fire department has reported that there are seven deaths and 40 injured. The witnesses said that a lot of smoke came out of the building with the explosion. The fire department has put 60 equipment and 160 people into the fire and the fire has been in effect in 20 minutes. However, only the basement of the building has sprinklers installed, and there is no anti-aircraft gun, so smoke spreads in the building and the damage to the people seems to have increased."
} |
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 d... | [
"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 ... | [
"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... | {
"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"
} | 32 | [
"A news broadcast featuring two anchors reporting on an arson attack at a law office in Daegu. The scene shows emergency vehicles and personnel at the site of the incident, with a red vehicle prominently displayed. The anchors discuss an investigation into a 50-year-old suspect, believed to have died at the scene, ... | 1534851387024359424 | A 50-year-old man, identified as A, is suspected of carrying out an arson attack on a law office in Daegu, resulting in the loss of seven lives, including six employees. The incident occurred due to A's dissatisfaction with the outcome of a trial related to a redevelopment project, in which he had invested and lost. A was seen entering the building, carrying a package wrapped in white cloth, and proceeding to the second floor, where he set the fire, blocking escape routes and causing the victims to succumb to smoke inhalation. The police investigation, led by reviewing CCTV footage, identified A as the suspect, and it is believed that he acted alone. The authorities are conducting a thorough investigation, gathering evidence and interviewing witnesses, to understand the full context and motive behind A's actions. The case may be closed due to lack of evidence if it is confirmed that A acted alone. The incident has resulted in significant damage to the law office and has caused widespread concern, with emergency responders and police officers actively involved in the response and investigation efforts. | {
"original": " 이번 대구 변호사 사무실 방화 용의자로 50대 남성 A씨를 지목하고 경찰이 수사에 착수했습니다. A씨는 화재 현장에서 숨졌는데 재판 결과에 불만을 품고 범행을 저지른 것으로 보입니다. 허성준 기자가 보도합니다. 한 남성이 흰색 천으로 감싼 물건을 들고 건물 계단을 올라갑니다. 2층으로 들어가면서 뭔가를 꺼내더니 곧이어 내부에서 시커먼 연기가 치솟습니다. 혼비백산한 사람들이 뛰쳐나오고 번쩍하는 섬광과 함께 CCTV도 작동을 멈춥니다. 남자가 들어간 곳은 불이 시작된 2층의 변호사 사무실. 이곳에서 직원 6명을 포함해 모두 7명이 한꺼번에 변을 당했습니다. 경찰은 CCTV 등을 확인해 방화 용의자로 50대 남성 A씨를 지목했습니다. 두 손으로 안은 모습이 확인됩니다. 경찰은 A씨가 사무실 출입구 쪽에서부터 인화성 물질을 뿌린 뒤 불을 붙인 것으로 보고 있습니다. 이 때문에 대피로가 막힌 직원들이 연기에 질식해 화를 당한 겁니다. A씨는 현장에서 숨졌고 분신했을 가능성도 제기됐습니다. 용의자 한 명의 단독 범행으로 일단은 추정되는데 만약에 이 용의자가 혼자 단독 범행으로 한 것이라면 그냥 사건은 공소권 없음으로 전결됩니다. A씨는 재개발과 관련된 투자금을 돌려달라고 소송을 냈다가 폐소하자 앙심을 품고 상대편 변호사 사무실에 범행을 저지른 것으로 알려졌습니다. 경찰은 용의자가 숨진 만큼 주변인 수사를 통해 사건 발생 경위와 범행 동기 등을 종합적으로 살필 방침입니다. YTN 허성준입니다. 감사합니다.",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "The police have launched an investigation into a 50-year-old man, identified as A, as the suspect in the arson attack on a law office in Daegu. A is believed to have died at the scene, and it appears that he committed the crime due to dissatisfaction with the outcome of a trial. According to reporter Heo Sung-jun, a man was seen carrying a package wrapped in white cloth and walking up the stairs of the building. As he entered the second floor, he pulled out an object, and soon, thick black smoke began to billow out from inside. People in a state of panic rushed out of the building, and the CCTV cameras stopped working due to the intense heat and flashing lights. The law office on the second floor, where the fire started, was where seven people, including six employees, lost their lives. The police identified A as the suspect after reviewing the CCTV footage, which showed him holding something in both hands. It is believed that A sprinkled flammable substances around the office entrance and set them on fire, blocking the escape routes and causing the employees to succumb to smoke inhalation. A died at the scene, and there is also a possibility that his body was severely burned. The police assume that A acted alone, but if that is the case, the investigation may be closed due to lack of evidence. A had reportedly filed a lawsuit to recover his investment in a redevelopment project, but after losing the case, he allegedly targeted the law office of the opposing party in the lawsuit. The police plan to conduct a thorough investigation, including interviewing witnesses and gathering evidence, to determine the circumstances surrounding the incident and the motive behind A's actions. This is YTN's Heo Sung-jun reporting.",
"translated_llm": "A man hides at the scene of the fire, but appears to have committed the crime out of dissatisfaction with the outcome of the trial. The reporter reports. A man carries a white ceiling covering around him and climbs the stairs. As he enters the 2nd floor, he takes out something, but immediately smoke rises from the inside of the building.",
"translated_whisper": " A man in his 50s pointed out A and the police started investigating. A seems to have committed a crime in the scene of the fire. Reporter Heo Sung-jun reports. A man takes a white cloth and goes up the stairs of the building. As he entered the second floor, he took out something and soon a black smoke came out from the inside. People who were in a daze came out and the CCTV stopped working with the sparkling ceiling. The place where the man entered was the second floor lawyer's office where the fire started. Here, all seven of the employees, including six employees, were changed at once. The police checked CCTV and pointed out the 50-year-old man A as a suspect. It was confirmed that the suspect was investigating the CCTV of the address. The police believe that A was lighting a fire after spraying chemical substances from the entrance of the office. This is why the employees who were blocked by the fire were angry at the smoke. A was suspected of hiding in the scene and likely to have been disfigured. It is assumed that the suspect is a single criminal, but if the suspect committed a single crime alone, the case is closed because there is no criminal record. When the A-sector filed a lawsuit to return investment money related to redevelopment, he was accused of committing a crime in the other side's lawyer's office with suspicion. The police will look at the case and the motive of the crime through the investigation of the surroundings as much as the suspect is hidden. This is YTN's Heo Sung-jun."
} |
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 d... | [
"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 ... | [
"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... | {
"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"
} | 32 | [
"A person is running in a hallway towards a door. The text on the screen reads \"침 침으로 뒤지는 물건 들고 건물 들어가는 방화 용의자,\" which suggests that the individual is suspected of arson while carrying an item.",
"A person is walking down a hallway carrying a red bag. The text on the screen reads \"현 천으로 뒤지는 물건 들고 건물 들어가는 방화 용의... | rzyM4QtZNNU | A suspected arsonist is shown entering a building, carrying a mysterious item, and proceeding to the second-floor office area. The individual climbs the stairs and enters the office, then exits and moves through the corridors near the elevator. The scene transitions to show the staircase and elevator area repeatedly, indicating a brief passage of time. Later, a person is seen running up the stairs, displaying urgency, and then descending with a focused posture as victims flee. The atmosphere becomes tense as the individual navigates through a smoky environment, emphasizing the need to escape quickly. The video then transitions to a news broadcast segment, featuring the KBS News logo and a call to action for viewers to subscribe. The overall context suggests a breaking news report about a suspected arson incident, capturing the actions of the suspect and the subsequent emergency response. | {
"original": " Terima kasih telah menonton",
"refined": "Thank you for watching",
"summarized": "The speaker is expressing gratitude to the audience for watching.",
"translated": "Thank you for watching.",
"translated_llm": "Thank you for watching.",
"translated_whisper": ""
} |
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 d... | [
"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 ... | [
"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... | {
"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"
} | 32 | [
"In the current frame, two news anchors are seated at a desk in a news studio, discussing a recent incident. The female anchor is wearing a blue blazer, and the male anchor is in a suit with a purple tie. They appear to be reporting on a fire that occurred at a lawyer's office building, which resulted in multiple f... | 5xiV1Z2MDX4 | A fire broke out on the 2nd floor of a 7-story lawyer's office building near the Daegu District Court in Daegu, resulting in 7 deaths and 46 injuries from smoke inhalation. The incident occurred around 11 am, with dozens of people evacuated in an emergency. Emergency services, including 50 fire trucks and 160 firefighters, responded quickly to extinguish the fire and rescue those trapped, bringing the blaze under control within 20 minutes. The police are investigating the cause of the fire, considering the possibility of arson due to reports of a client expressing dissatisfaction around the time the fire started. The video depicts the aftermath of the fire, with emergency responders, including firefighters and police officers, actively engaged in rescue and investigation efforts outside the building. The scenes show significant damage to the building, with ladders positioned against it, indicating ongoing rescue or inspection efforts. The emergency response and investigation into the cause of the fire continue, with a focus on determining whether the fire was intentionally set. | {
"original": " 오늘 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 들렸다는 신고가 접수됐습니다. 불이 나자 소방차 50대와 소방인력 160명이 동원돼 불을 끄고 입주자들을 구조했고 이번 화재는 20여 분 만에 진화됐습니다 경찰은 화재 발생 당시 의뢰인이 불만을 제기한 정황이 있었다는 점 등으로 미뤄 방화 여부를 조사하고 있습니다",
"refined": "At 11 a.m. today, a fire broke out on the 2nd floor of a 7-story law office building near Daegu District Court in Daegu City's Suseong-gu, Beomheo-dong. Seven people died, and 46 suffered injuries from smoke inhalation. Dozens of people in the building were evacuated in an emergency. The deceased were five men and two women. When the fire was reported to 119, witnesses saw black smoke and heard an explosion on the 2nd floor. Firefighters responded with 50 vehicles and 160 personnel, extinguishing the fire and rescuing occupants within 20 minutes. The police are investigating whether the fire was an act of arson, citing a complaint filed by a client at the time of the incident.",
"summarized": "A fire broke out at a 7-story office building near the Daegu District Court in South Korea, resulting in 7 deaths, with 5 men and 2 women confirmed deceased. The fire also injured 46 people and led to the emergency evacuation of dozens of others. The fire was extinguished within 20 minutes with the help of 50 fire trucks and 160 firefighters. The police are currently investigating the cause of the fire, including the possibility of arson, as there were reports of a complaint from a client at the time of the incident.",
"translated": "At around 11 am today, a fire broke out on the 2nd floor of a 7-story lawyer's office building near the Daegu District Court in Suseong-gu, Bumseo-dong, Daegu. The fire resulted in 7 deaths, with 5 men and 2 women reported among the fatalities, and 46 people were injured from inhaling smoke. Dozens of people inside the building were evacuated in an emergency. At the time, a report was made to the emergency services, stating that black smoke was visible and an explosion was heard coming from the 2nd floor. In response, 50 fire trucks and 160 firefighters were deployed to extinguish the fire and rescue those trapped, managing to bring the blaze under control within 20 minutes. The police are investigating the cause of the fire, including the possibility of arson, as there were reports of a client expressing dissatisfaction around the time the fire broke out.",
"translated_llm": "At around 11 am today, a fire broke out in the 7 storey lawyer's office building on the second floor near the Suwon-gu Gangar-east District Court. Seven people hid and 46 people were injured after drinking the smoke. Dozens of people inside were evacuated.",
"translated_whisper": " At 11 a.m. today, a fire broke out in the 2nd floor of Daegu District Court's 7-story office building near Daegu City, Suseong-gu, Beomheo-dong. 7 people were breathing and 46 people were injured, including smoking. Also, dozens of people in the building were urgently evacuated. The number of deaths was known to be 5 men and 2 women. At the time of the 119, a report was received that black smoke was seen and explosion was heard on the 2nd floor. When the fire broke out, 50 firefighters and 160 firefighters were mobilized to turn off the fire and rescue the residents. This fire has been evolved in 20 minutes. The police are investigating the prevention of fire by pushing the fact that the client had a complaint at the time of the fire."
} |
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 hitti... | [
"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, Sout... | [
"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"... | {
"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"
} | 32 | [
"The image depicts a nighttime scene with heavy rain, creating a misty and gloomy atmosphere. Streetlights illuminate the surroundings, but visibility is low due to the rain and darkness. Trees and buildings are visible, partially obscured by the rain. There is no indication of seating or people, and the focus is o... | 1566887846916792320 | The video depicts a nighttime scene with heavy rain, creating a misty and gloomy atmosphere. The setting appears to be a street with trees and buildings, illuminated by streetlights that struggle to penetrate the darkness and rain. For most of the video, the scene is static, focusing on the adverse weather conditions. However, at some points, a car with its headlights on is visible driving along the wet road, adding a sense of movement to the otherwise isolated scene. A red traffic light is also visible in the foreground of some frames, adding a hint of color and structure. The overall atmosphere is one of isolation and challenging conditions, highlighting the difficulties of navigating through such heavy rain at night. There is no indication of people or seating areas, contrary to the mention of "The Kizukawa Seating" in the audio transcript, which does not seem to correlate with the visual content of the video. | {
"original": " The Kizukawa Seating",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "The Kizukawa Seating",
"translated_llm": "The Kizukawa Seating",
"translated_whisper": " The public at the public at the public board board board board board board"
} |
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 hitti... | [
"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, Sout... | [
"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"... | {
"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"
} | 32 | [
"In this dark and shadowy frame, the focus is on a dense cluster of trees in the foreground, silhouetted against the night sky. The scene appears to be set in an outdoor area, possibly a park or a garden, with power lines faintly visible in the background. The overall ambiance is quiet and still, creating a serene ... | 1566774984525025281 | The video depicts a dark and introspective scene, set in an outdoor area at night, with a dense cluster of trees silhouetted against the sky and power lines faintly visible in the background. The atmosphere is quiet and still, creating a contemplative mood that persists throughout the video. The narrative revolves around personal reflection and introspection, with the speaker pondering their stage and experiences, drawing a comparison to IU. The speaker expresses a sense of wonder and questioning about their individuality, asking if they are the only one like IU, indicating a deepening self-reflection and search for uniqueness within their journey. The video maintains a consistent theme of personal introspection, emphasizing the speaker's thoughts and emotions as they navigate their path, influenced by their perceptions of IU. | {
"original": " 이번 해에 빛을 띠는 이 무대는 저의 첫 번째 무대를 생각해보려 했는데 아직도 아이유가 한 명은 아닌지",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "This year's stage that shines, I tried to think of my first stage, but I still wonder if I'm the only one like IU.",
"translated_llm": "This is my first stage this year, but I still don't have one.",
"translated_whisper": " Translation provided by fans!"
} |
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 hitti... | [
"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, Sout... | [
"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"... | {
"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"
} | 32 | [
"The current frame captures a scene inside an indoor facility, likely an ice rink, that has partially flooded with water. Several individuals are seen in the water, some standing and others sitting. Onlookers are gathered behind a railing, observing the situation. One person is holding a phone, presumably recording... | 1567139282846089217 | A group of individuals, including rescue workers or assessors in safety gear, are navigating and managing a flooded indoor area, likely a parking structure, as indicated by the "PARKING OUT" sign. The flood covers the floor, making movement difficult, and the individuals are working together, using equipment such as flashlights and a yellow machine, possibly a pump, to clear the water. The scene transitions to the group being outside, continuing their efforts to manage the flood situation, with ongoing coordination and teamwork evident. Throughout the video, the audio features a repetitive phrase, suggesting a casual or ongoing situation, as the individuals persist in their efforts to mitigate the impact of the flood. | {
"original": " . . . . . . . . . . . . . . 우리 아들 흔하게 가죠 흔하게 흔하게 흔하게 나왔다 나왔다 흔하게 흔하게 흔하게 흔하게",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Our son is coming out commonly, commonly, commonly, commonly. Let's go, commonly, commonly, commonly, commonly.",
"translated_llm": ". . . . . . . . . . . . . . . . අපේ පුතා 흔하게 가죠 흔적으로 흔적으로 흔적으로 나왔어요 흔적으로 흔적으로 흔적으로 흔적으로",
"translated_whisper": " The third day in Seoul On the 4th day of the national team's first game The first time the team had a 2-on-4 show at a local venue The first match ended on a special day Where the first team played, but the last one got cancelled I can't believe it. There was a first-team match in the first game There was a chance that there was no game I'm going to my son. Let's go to the cinema. The cinema. The cinema. I'm here. I'm here."
} |
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 hitti... | [
"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, Sout... | [
"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"... | {
"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"
} | 32 | [
"In the current frame of the video, two news anchors are seated at a news desk in a studio. The male anchor on the left appears to be speaking, gesturing with his hands to emphasize points, while the male anchor on the right holds papers and listens attentively. Behind them, a large screen displays turbulent ocean ... | 1566015450982494209 | A news broadcast discusses the impending Typhoon "Maemi", which is expected to bring record-breaking wind speeds of up to 60 meters per second, posing a significant threat to various regions in South Korea. The typhoon's potential impact is compared to a historical typhoon that hit Busan in 2003, causing the collapse of a Goliath crane due to winds of 40 to 50 meters per second. The news anchors emphasize the catastrophic potential of the typhoon, particularly in areas such as Jeju Island, South Gyeongsang Province, and the eastern coast of Gyeongsang Province, where winds over 50 meters per second can cause widespread destruction. The broadcast also highlights the risks in Seoul, where expected rainfall of 100 to 300 millimeters combined with winds over 15 meters per second can lead to significant damage and loss of life. The Meteorological Office warns of a possible storm surge if the typhoon approaches during high tide, and the news anchors stress the need for caution and preparedness in affected areas. The video shows images of empty streets, stormy weather, and news anchors discussing the typhoon's impact, emphasizing the urgency and potential dangers of the situation. | {
"original": " 초속 한 54m가 넘어가면 건물도 붕괴될 정도가. 그렇습니다. 달리는 열차가 탈선될 수도 있고요. 또 구조물이 무너질 수도 있습니다. 대표적인 예가 2003년도에 매미가 왔을 때 부산 지역 같은 경우는 초속 40m에서 50m의 바람이 불었습니다. 그럴 때 골리앗 크레인이 무너졌습니다. 그러니까 그 정도로 철제 구조물을 붕괴시킬 정도로 강한 바람이 초속 50m 이상의 바람이거든요. 그런데 초속 60m가 분다고 생각을 해보십시오. 그게 제주도 산간지역만이 아니라 경남 해안지역, 남해안지역과 경상동해안지역까지 그런 바람이 분다고 한다면 정말 엄청난 피해가 날 수밖에 없는 상황이고. 그러네요. 그보다는 조금 덜하지만 영동지방, 동해안지역과 울릉도 독도 쪽에도 30에서 40m의 초속 강풍이 불 겁니다. 서울은 그러면 괜찮으냐. 서울은 100에서 300mm의 비가 내린다고 예상했는데 거기에 바람도 초속 15m 이상입니다. 그 정도는 어느 정도인가요? 간판이 날아갈 수 있는 거죠. 그렇죠. 간판이 날아갈 수 있는 거고 불안전한 시설물은 얼마 전 떨어지면서 인명피해까지 날 수 있는 상황입니다. 그러니까 비와 바람을 동시에 동반한 굉장히 무서운 태풍이다라는 게 여기에서 여실히 드러나는 거죠.",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "If the wind speed exceeds 54 meters per second, it can cause buildings to collapse. The running train may derail, and structures may also collapse. A notable example is the 2003 typhoon that hit the Busan area, where winds of 40 to 50 meters per second were recorded, resulting in the collapse of a Goliath crane. This demonstrates that winds of over 50 meters per second can be strong enough to destroy steel structures. Now, imagine winds of 60 meters per second. If such strong winds were to hit not only the mountainous areas of Jeju Island but also the coastal regions of South Gyeongsang Province, the southern coast, and the eastern coast of Gyeongsang Province, the damage would be catastrophic. Although somewhat less severe, the Yeongdong region, the East Sea coast, and the Ulleungdo and Dokdo areas can expect strong winds of 30 to 40 meters per second. As for Seoul, while it may seem relatively safe, the expected rainfall of 100 to 300 millimeters, combined with winds of over 15 meters per second, poses a significant threat. Winds of this speed can sweep away signs and potentially cause unstable structures to collapse, leading to loss of life. This highlights the ferocity of the typhoon, which brings both heavy rain and strong winds.",
"translated_llm": "If the speed exceeds 54 m/s, the building can collapse. Yes. The running trains can be derailed. The structures can also collapse. A typical example is that in 2003, when the tsunami hit the Busan area, there was a wind of 40 to 50 m/s. Then the Goliath crane collapsed. So there is a wind strong enough to collapse the iron structure at a speed of more than 50 m/s.",
"translated_whisper": " If the wind blows over 54m per second, the buildings will collapse. That's right. The train that runs may be decommissioned. The structure may collapse. A typical example is when the cicada came in 2003, the wind in Busan was 40m to 50m per second. Then the Goliath Crane collapsed. So the wind is stronger than 50m per second to the extent that it collapses the steel structure. But think about blowing 60m per second. If you look at the wind in the south and east coast of Gyeongsang-do, it is a situation where there is no choice but to be damaged. It is a little less than that, but the wind will be strong in the East Sea and the Ulleungdo Dokdo region. Seoul is 100-300mm above sea level, but the wind is 15m above sea level. How much is that? The sign can fly. The sign can fly, and the unstable facilities can fall soon and cause a fatal failure. So it's a very scary storm that accompanies rain and wind at the same time."
} |
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 hitti... | [
"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, Sout... | [
"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"... | {
"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"
} | 32 | [
"The current frame shows a significant landslide with a large section of earth and debris cascading down a hillside. In the background, there are several high-rise buildings, indicating that this event is occurring in an urban area. A modern building is visible at the bottom of the frame, seemingly unaffected by th... | 1566948535899041792 | A landslide occurred in an urban area, causing significant disruption to the landscape. The event was captured in a series of frames showing the aftermath of the landslide, with a large section of earth and debris sliding down a hillside near high-rise buildings. Despite the proximity to the buildings, they remained intact and unaffected, along with a modern building and a parked car at the bottom of the hill. The surrounding trees and vegetation were impacted, with visible damage and uprooted trees. The overcast sky suggested recent adverse weather conditions that likely contributed to the landslide. The audio transcript "Wow, I didn't miss school, did I?" reflected a sense of relief or disbelief from someone observing the scene, highlighting the dramatic and unexpected nature of the event. Throughout the frames, the landslide's impact on the terrain was evident, with the debris and earth settling and causing significant changes to the landscape. The high-rise buildings, modern building, and parked car remained undamaged throughout, while the surrounding environment continued to show signs of damage from the landslide. | {
"original": " 우와, 학교 못나간건 아니가. 어!",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Wow, I didn't miss school, did I?",
"translated_llm": "No, I didn't go to school.",
"translated_whisper": " Wow... Oh, isn't it a school gate? Oh.. Ah !!!!!!"
} |
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",
"Pe... | [
"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... | [
"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... | {
"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"
} | 32 | [
"The current frame shows a rocky coastline with turbulent waves crashing against the rocks. The text on the image in Korean reads, \"One can't go back by the same path; it's truly a journey of frustration.\" This likely reflects a theme of the video, possibly discussing challenges or the idea that life paths are ir... | QJY4XlSNVf8 | The video depicts a severe weather event, specifically typhoons and heavy rainfall, affecting various regions, including China's Jizang and Hainan provinces, and potentially impacting the metropolitan area. The footage shows turbulent waves, flooded streets, and rescue efforts, with text overlays providing information on the typhoons' influence and the resulting damage. The video also shifts to an indoor conference room setting, where officials or experts discuss and analyze the storm's impact, displaying maps and graphs on a large projection screen. The meeting appears to be a formal, coordinated response to the weather event. The video concludes with a logo and subscription prompts from Yonhap News Agency, encouraging viewers to subscribe. Overall, the video showcases the challenges posed by severe weather conditions and the efforts of authorities to monitor, respond to, and prepare for such events. | {
"original": " Thank you. Thank you.",
"refined": "Thank you.",
"summarized": "Not Available",
"translated": "Thank you, thank you.",
"translated_llm": "Thank you, thank you.",
"translated_whisper": " The The The The The The The The The The The The The The The you"
} |
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",
"Pe... | [
"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... | [
"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... | {
"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"
} | 32 | [
"In the current frame of the video, two news anchors are seated in a news studio. The female anchor is on the left wearing a maroon blazer and a light-colored blouse, while the male anchor on the right is dressed in a dark suit with a white shirt and a blue tie. Behind them, multiple monitors display various images... | ZFR-SBQwtTw | The video depicts a news broadcast covering the formation and potential impact of Typhoon Lingling on the Korean Peninsula. The broadcast features two news anchors discussing the latest developments regarding the typhoon, with footage of strong winds, rough seas, and preparations for the storm's arrival. The typhoon is shown to be causing significant disruptions, with palm trees swaying violently, waves crashing against shorelines, and people seeking shelter. The news anchors provide updates on the typhoon's path, which is expected to move north towards the Korean Peninsula in the latter half of the week, potentially affecting the region by Friday morning. The video also shows scenes of people securing boats, covering street vendor carts, and taking other precautions to prepare for the storm. Overall, the video provides a comprehensive overview of the situation, highlighting the potential risks and impacts of Typhoon Lingling on the Korean Peninsula. | {
"original": " 오늘 오전 9시쯤 필리핀 동쪽 해상에서 13호 태풍 링링이 발생했습니다. 강풍 반경 200km의 소형 태풍인 링링은 계속 북상해서 수요일쯤 타이완 동쪽 해상에 접근하겠습니다. 태풍 링링은 금요일 오전쯤 제주도 서귀포 남서쪽 390km 해상까지 진출하겠고, 이후 서해상을 향할 걸로 보입니다. 기상청은 태풍 발생 초기라 경로가 유동적이지만 주 후반부터 한반도 부근으로 북상할 가능성이 크다고 내다봤습니다.",
"refined": "Typhoon Lingling, with a radius of 200 kilometers, occurred in the eastern waters of the Philippines around 9 a.m. today. The small typhoon will continue to move north and approach the eastern waters of Taiwan by Wednesday. Typhoon Lingling is expected to reach the southwest of Seogwipo, Jeju Island, around Friday morning, and then head towards the Yellow Sea. The Korea Meteorological Administration noted that while the typhoon's path is still uncertain at this early stage, it is likely to move north towards the Korean Peninsula from the second half of the week.",
"summarized": "Typhoon Lingling occurred in the Philippine East Sea at 9 am today, with a strong wind radius of 200 km. The typhoon is expected to approach the Taiwan East Sea by Wednesday and move towards Jeju Island's southwestern waters by Friday morning. It is then predicted to head towards the west sea. Although the typhoon's path is currently uncertain, it is likely to move towards the Korean peninsula in the latter half of the week.",
"translated": "At around 9 am today, the 13th typhoon, Lingling, formed in the eastern waters of the Philippines. As a small typhoon with a strong wind radius of 200km, Lingling is expected to continue moving north and approach the eastern waters of Taiwan by Wednesday. By Friday morning, Typhoon Lingling is forecasted to reach a point 390km southwest of Seogwipo, Jeju Island, and is then expected to head towards the Yellow Sea. The Korea Meteorological Administration notes that while the typhoon's path is subject to change due to its early stage, it is likely to move north towards the Korean Peninsula in the latter half of the week.",
"translated_llm": "Today at 9 am, 13th Typhoon Ringling occurred in the eastern sea of the Philippines. The strong winds of 200km radius will continue northward and approach the eastern sea of Taiwan by Wednesday. The typhoon ringling will enter the sea of the southwest of the country of Singapore at 390km on Friday morning, and then appear to be heading towards the western sea.",
"translated_whisper": " A 13-minute typhoon, Lingling, occurred at the Philippine East Sea Sea on 9 a.m. today. Lingling, a small typhoon 200 kilometers in the radius of the river, will continue to approach the Taiwan East Sea Sea on Wednesday from the north. The typhoon Lingling will advance to the southern west of Jeju Island, Seogwipo, around Friday morning, and will then head to the west sea. The weather office said that the route is dynamic because it is early for a typhoon to occur, but it is likely to rise to the It is likely to go up."
} |
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",
"Pe... | [
"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... | [
"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... | {
"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"
} | 32 | [
"The current frame depicts a weather map showing a cyclone track in the Philippines. The text in Korean on the image translates to \"Cyclone forming in the eastern Philippines, with a 13-hour movement forecast 'Lingling'.\" The map highlights a specific area in the Philippines where the cyclone is currently located... | 8-DXe4qYo2E | The video depicts a severe weather situation involving typhoons, specifically Cyclone Lingling and Typhoon Melor, affecting the Philippines, South Korea, and potentially Japan. The frames show weather maps tracking the cyclones' movements, forecasts of their paths, and the potential impact areas. The video then transitions to show the actual effects of the typhoon, with scenes of powerful waves crashing against shorelines, strong winds, and heavy rain causing flooding in urban areas. News reports are interspersed, providing updates on the typhoons' locations, speeds, and expected trajectories. The video concludes with an animated character from the Yonhap News Agency encouraging viewers to subscribe, indicating the source of the weather updates and news reports. Overall, the video documents the progression and impact of severe typhoons in the region, emphasizing the importance of weather forecasting and preparedness. | {
"original": " To be continued... Thank you.",
"refined": "To be continued... Thank you.",
"summarized": "Not Available",
"translated": "To be continued... thank you.",
"translated_llm": "To be continued... thank you.",
"translated_whisper": " The The The The The The The The The The The The The The To be continued... Thank you."
} |
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",
"Pe... | [
"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... | [
"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... | {
"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"
} | 32 | [
"The image shows a vehicle driving down a street during a storm, with strong winds causing trees to sway and debris to fly. The Korean text suggests that it is capturing the intensity of a powerful storm, potentially a typhoon.",
"The image shows a street scene during a storm, with strong winds causing a tree to ... | taOktWIvk20 | The video depicts the impact of Typhoon Chanthu on various locations, primarily in South Korea. The storm brings strong winds and heavy rain, causing damage to infrastructure, including fallen utility poles, debris-filled streets, and disrupted maritime activities. The footage shows hazardous conditions at docks, with fishing boats swaying violently and a person struggling in the water. The typhoon also affects daily life, with traffic stopped on highways due to strong winds and low visibility. Authorities, including police officers, are seen managing traffic and ensuring safety during the storm. The video concludes with a transition to a news segment, encouraging viewer engagement with the Yonhap News Agency. Overall, the video highlights the severe weather conditions and the efforts to manage and recover from the typhoon's impact. | {
"original": " 강원도에 소강기 사치장이 발견 주차장의 전파가 시작됐습니다 교체점으로부터 배경이 전파가 공동화됐습니다 지하철을 가기 전에 전파가 공동화됐습니다 교체업관에 있는 방문해 주셔야지요 비행기가 불이 잘 틀리니까 감사합니다.",
"refined": "I couldn't refine the translation with reasonable accuracy.",
"summarized": "The speaker appears to be giving directions and describing a location, but the content is unclear and inconsistent between the original and translated versions.",
"translated": "A radio broadcast has started from a parking lot in Gangwon-do, spreading from a relay point, and was relayed before taking the subway, so please visit the replacement observatory, as the plane's broadcast is not coming in well, thank you.",
"translated_llm": "The radio station is now on the ground, and the station is in the parking lot.",
"translated_whisper": " The most famous train station of Korea. Go straight on the left, then turn right. Just a few steps toward Yeouido, We found the KBF-100H train station. Going to see KBF-100H now. I saw it this morning, and it looks so good."
} |
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",
"Pe... | [
"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... | [
"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... | {
"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"
} | 32 | [
"The image shows an urban scene with a fallen tree in the foreground, uprooted and lying across the sidewalk. There are bicycles parked nearby, one of which has a flag with Korean text attached to it. The background includes various storefronts and buildings, indicating the aftermath of a storm or heavy wind.",
"... | t1uezBvD9U8 | The video depicts the aftermath of a severe weather event, likely a storm, that caused significant damage to an urban area. The scene shows a large, uprooted tree lying across a city street, blocking traffic and affecting urban infrastructure. Debris is scattered around, and storefronts, including a "Paris Baguette" shop, are visible in the background. The fallen tree and debris on the street highlight the extent of the disruption in the city. As the video progresses, it shows the ongoing efforts to manage the situation, with people walking past the fallen tree and debris, emphasizing the disruption to daily life. The scene also shows a damaged section of a wall, a staircase, and a slanted black board with circular objects connected by strings, possibly part of a makeshift installation. The video also depicts a large, triangular sign with the letters "47" prominently displayed, surrounded by caution tape, indicating ongoing recovery efforts. The scene shifts to a fenced-off area, possibly a sports facility, with caution tape and a small building inside, suggesting restricted access due to potential hazards. The video then shows a park-like area with lush green trees and a serene atmosphere, indicating a possible return to normalcy. However, the scene soon shifts to a fenced area with damaged netting and a leaning fence, suggesting neglect or abandonment. The video ends with a scene of a collapsed structure, possibly a building or billboard, with a man in an orange shirt and helmet inspecting the damage, indicating an effort to assess the situation and possibly take steps towards recovery or safety. Throughout the video, the dramatic music and subtitles by Nicolai Winther maintain a somber tone, underscoring the ongoing impact of the weather event on the community. | {
"original": " Dramatisk musikk Undertexter av Nicolai Winther",
"refined": "Dramatic music subtitles by Nicolai Winther",
"summarized": "The speaker is introducing a video tutorial on a topic that will be demonstrated, but the topic itself is not specified in the given transcripts.",
"translated": "Dramatic music Subtitles by Nicolai Winther",
"translated_llm": "Dramatisk musik Subtitled by Nicolai Winther",
"translated_whisper": " Hello! In this video, I will show you how to make a"
} |
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",
"H... | [
"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",
"Floo... | [
"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 f... | {
"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"
} | 32 | [
"The video frame shows a view from inside a vehicle, looking through a slightly foggy windshield at a street scene. The scene outside appears to be in low visibility conditions, possibly due to rain or mist. The street is lined with buildings, and a few pedestrians can be seen walking along the sidewalk. The windsh... | 1032900418777407489 | A vehicle is stationary or moving very slowly in a street lined with buildings on a rainy or misty day. The windshield wipers are continuously active, attempting to clear the moisture from the windshield to improve visibility. Pedestrians and other vehicles are occasionally visible on the street, despite the low visibility. At some point, a person is seen outside, waving or signaling, but they eventually disappear from view. The interior of the vehicle, including the dashboard and driver's seat, remains visible throughout, with a red object hanging from the rearview mirror. The video primarily focuses on the view through the windshield, showcasing the efforts of the windshield wipers to maintain visibility during the wet conditions. | {
"original": "",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Not Available",
"translated_llm": "- What?",
"translated_whisper": ""
} |
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",
"H... | [
"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",
"Floo... | [
"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 f... | {
"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"
} | 32 | [
"In the current frame, the video captures a street scene where palm trees are swaying violently in the wind. The street is wet, suggesting recent or ongoing rainfall. There are several cars on the road, moving cautiously, and debris, possibly from a fallen tree, is visible on the road. The turbulent weather conditi... | 1032211760315420675 | The video depicts a street scene affected by a storm, with strong winds and heavy rain causing debris, including palm fronds and branches, to obstruct the road. The storm's impact is evident, with fallen trees blocking the road and leaning towards houses in a residential area. Despite the hazardous conditions, some cars are seen moving cautiously through the debris-strewn roads. The video shows the aftermath of the storm, with wet streets, swaying trees, and a calm, serene atmosphere. The scene is devoid of people, emphasizing the quiet and stillness following the storm. The video does not appear to show any direct celebration or acknowledgment of the mentioned birthday, instead focusing on the storm's effects on the environment. | {
"original": " Good morning. Hello everyone Today is the birthday of the first-ever A.V. of A.V. I'm very sorry",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Good morning. Hello everyone. Today is the birthday of the first-ever A.V. of A.V. I'm very sorry.",
"translated_llm": "Today is the birthday of the first ever A.V. of A.V.",
"translated_whisper": " All of the"
} |
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",
"H... | [
"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",
"Floo... | [
"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 f... | {
"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"
} | 32 | [
"The image is a weather map showing a typhoon over the East China Sea, with significant wind patterns and pressure systems visible. The map includes temperature readings and city names in both Korean and English, indicating regions in North and South Korea, as well as parts of China and Japan. The typhoon's eye is ... | 1032540315515682816 | The video appears to be a weather forecast or update, focusing on the East China Sea and the Sea of Japan, with a particular emphasis on the Korean Peninsula, China, and Japan. The video displays a series of weather maps, initially showing one typhoon and later two typhoons, with their paths, intensities, and potential impacts on the mentioned regions being closely monitored. The maps include temperature readings and city names in both Korean and English, indicating the areas that could be affected by the typhoons. As the video progresses, the focus shifts towards a high-pressure system with warm temperatures in the southern regions of China and Japan, suggesting a trend towards warmer conditions. Later frames introduce a low-pressure system near the Korean Peninsula, which appears to be intensifying, potentially leading to changes in weather conditions. Throughout the video, the geographical context is maintained, with the text "NORTH KOREA" and "SOUTH KOREA" visible in some frames. The overall context suggests a weather update or forecast, aiming to inform viewers about the current and potential future weather conditions in the region. | {
"original": "",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Not Available",
"translated_llm": "- What?",
"translated_whisper": ""
} |
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",
"H... | [
"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",
"Floo... | [
"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 f... | {
"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"
} | 32 | [
"A large tree with lush green leaves stands in a muddy area, likely affected by rain or flooding, with some debris visible on the ground. The tree is swaying slightly, suggesting windy conditions.",
"The video captures a large tree with lush green leaves standing in a muddy area, likely affected by rain or floodi... | 1032471495958618112 | The video depicts a natural setting with a large, leafy tree standing in a muddy and wet area, indicating recent or ongoing rainfall. The tree is consistently swaying slightly throughout the video, suggesting strong and persistent windy conditions. The surrounding area is filled with other trees and vegetation, contributing to a natural and somewhat turbulent atmosphere. The scene remains relatively static, with the primary focus on the swaying tree and the effects of the wind and rainfall on the environment. The video does not appear to show any specific actions or participants beyond the natural movements of the tree and the environmental conditions. Overall, the video captures a serene yet turbulent natural scene, emphasizing the impact of weather on the environment. | {
"original": "",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Not Available",
"translated_llm": "- What?",
"translated_whisper": ""
} |
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",
"H... | [
"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",
"Floo... | [
"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 f... | {
"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"
} | 32 | [
"A severe storm is affecting the area, as indicated by the text in the image. Strong winds are visible, with debris being blown around and a car partially obscured by the storm. The location is specified as Amami Ōshima, and the time is just before 6 PM. The scene suggests strong winds and possibly heavy rain, cons... | 1032234901897015298 | A severe storm, identified as Typhoon 19, is affecting the area of Amami Ōshima. The storm brings strong winds and heavy rain, reducing visibility and causing flooding in the area. A yellow vehicle is seen navigating through the stormy conditions, with debris and rain being blown around, emphasizing the strength of the storm. The text overlays throughout the frames indicate the location as Amami Ōshima, the time as around 6-7 PM, and mention the storm's "extremely strong power" and "severe warning." The situation remains challenging due to the typhoon's impact, with low visibility, strong winds, and flooding persisting throughout the video. The video depicts the ongoing severe weather conditions, highlighting the significant impact of Typhoon 19 on the area. | {
"original": "",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Not Available",
"translated_llm": "- What?",
"translated_whisper": ""
} |
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",
"H... | [
"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",
"Floo... | [
"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 f... | {
"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"
} | 32 | [
"The video frame shows a news anchor reporting on a weather event, specifically discussing the impact of a typhoon. The background features a weather map depicting wind patterns and the path of the typhoon. The text \"최혜림\" on the screen indicates the name of the news anchor. The context provided by the ASR transcr... | _V6HiUPBShQ | The video reports on Typhoon Solik, which caused damage in Jeju and inland areas of the Korean Peninsula due to its rapid weakening. The typhoon, initially strong with a clear eye, was slowed down by Typhoon Cimarron, causing it to linger in the seas west of Jeju for 24 hours. During this time, Solik released its energy and moisture, leading to heavy rain in Jeju. The prolonged stay over the cold waters and the influence of a strong west wind disrupted the typhoon's structure, causing it to rapidly lose strength. The typhoon's eye disappeared, and its shape became distorted. Solik made landfall in Mokpo, at the southwestern tip of the Korean Peninsula, and quickly lost its strength as it was no longer able to draw in moisture, eventually moving out of the peninsula. The report explains that the interaction between Typhoon Solik and Typhoon Cimarron, along with the influence of the North Pacific high-pressure system, contributed to Solik's rapid weakening. The video features a news anchor discussing the typhoon's impact, satellite images of the typhoons, and scenes of the aftermath, including turbulent ocean waves, swollen rivers, and damaged infrastructure. Experts, such as Mun Il-joo from the Jeju National University Typhoon Research Center, provide insights into the scientific factors behind the typhoon's behavior. The report concludes by highlighting the unpredictability of weather interactions and their significant impact on the region. | {
"original": " 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. 또렷한 태풍의 눈을 유지한 채 기세 등등 올라오던 솔릭이 제주 서해상에 붙잡힌 건 20호 태풍 시마론의 영향이 큽니다. 솔릭은 북태평양 고기압 가장자리를 타고 움직였는데 시마론이 그 사이로 들어오면서 고기압을 동쪽으로 밀어냈습니다. 여기에 시마론에서 부는 바람이 솔릭을 남쪽으로 잡아당기며 북상 속도를 늦췄습니다. 이러다 보니 사람 걸음 정도로 속도가 떨어지기도 하며 솔릭은 제주 서해상에 24시간이나 머뭅니다. 그동안 제주에 폭우를 뿌리며 에너지와 수증기를 쏟아버렸습니다. 쏟아버렸습니다. 한 곳에 오래 머물다 보니 해수면 아래 찬물까지 끓어올리게 됐고 그러면서 급격히 힘을 잃은 겁니다. 여기에 일찍 올라갔으면 못 만났을 강한 편서풍이 내려오면서 태풍에 경고했던 구조를 깨뜨렸습니다. 깨뜨렸습니다. 결국 이 시기에 태풍의 눈은 사라졌고 태풍의 모양은 완전히 흐트러집니다. 이렇게 약화된 솔릭은 한반도 서남단인 목포로 상륙하면서 더 이상 수증기를 공급받지 못한 채 급격히 위력을 잃고 한반도를 빠져나갔습니다. SBS 엄민재입니다. SBS 엄민재입니다.",
"refined": "I told you about the typhoon damage. The damage was concentrated in Jeju and some inland areas because Solik's power dropped rapidly, unlike the forecast. The typhoon, which caused anxiety with strong winds over 100 km/h and heavy rain, weakened quickly. Reporter Eom Min-jae explains why. Solik, which maintained a clear eye, was slowed down by the 20th typhoon, Cimarron. Solik was moving along the edge of the North Pacific high-pressure system, but Cimarron intervened, pushing the high-pressure system eastward. The winds from Cimarron also pulled Solik southward, slowing its northward speed. As a result, Solik's speed dropped, and it stayed in the Jeju Sea for 24 hours, pouring heavy rain and energy over Jeju. This prolonged stay led to the upwelling of cold seawater, causing Solik to lose its strength rapidly. The strong westerly winds that descended at this time disrupted the typhoon's structure, which had been a cause for concern. Consequently, the typhoon's eye disappeared, and its shape became distorted. The weakened Solik made landfall in Mokpo, at the southwestern tip of the Korean Peninsula, and lost its power as it could no longer draw energy from the sea.",
"summarized": "A typhoon, Solic, caused damage mainly in Jeju and some inland areas due to its sudden loss of power, contrary to forecasts. This was largely influenced by the 20th typhoon, Shimaron, which disrupted Solic's path and pulled it south, slowing its northward movement. As a result, Solic lost energy and water vapor while staying over the Jeju Seohae for 24 hours. Its strength was further diminished as it encountered cold ocean water and strong winds, eventually breaking its structure and losing its power. The weakened storm made landfall in Mokpo, South Korea, and quickly lost power and escaped the peninsula.",
"translated": "I reported on the damage caused by the typhoon. The damage was concentrated in Jeju and some inland areas because the typhoon's strength decreased rapidly, contrary to the forecast. The typhoon, which had brought strong winds of over 100 km/h and heavy rain, causing widespread concern, weakened so quickly. Reporter Eom Min-jae explains why. Typhoon Solik, which had maintained its strong momentum with a clear eye, was slowed down in the seas west of Jeju due to the influence of Typhoon Cimarron. Solik was moving along the edge of the North Pacific high-pressure system, but Cimarron intervened, pushing the high pressure to the east. The winds from Cimarron also pulled Solik southward, slowing its northward movement. As a result, Solik's speed dropped to a pace of about a person's walking speed, and it lingered in the seas west of Jeju for 24 hours. During this time, it poured heavy rain on Jeju, releasing its energy and moisture. As it stayed in one place for an extended period, the cold water from beneath the sea surface rose, causing Solik to rapidly lose strength. Additionally, a strong west wind that would not have been encountered if Solik had moved earlier came down and disrupted the typhoon's structure. Consequently, the typhoon's eye disappeared, and its shape became completely distorted. The weakened Solik made landfall in Mokpo, at the southwestern tip of the Korean Peninsula, and as it was no longer able to draw in moisture, it rapidly lost its strength and moved out of the peninsula. This is Eom Min-jae reporting for SBS.",
"translated_llm": "The storm was reported to have caused significant damage in Jeju and parts of the mainland. The damage was concentrated because the force of Solrik dropped sharply, unlike the forecast. The storm, which was over 100 km per hour, was accompanied by strong winds and rain, and caused everyone to feel nervous. The storm weakened so quickly.",
"translated_whisper": " I told you about the typhoon damage. The damage to Jeju and some of the mainland was concentrated because the power of Solic was rapidly reduced, unlike the forecast. Why was the typhoon, which was nervous with the strong winds and rain, over 100 kilometers per hour, so weak? Reporter Eom Min-jae tells you. Solic, which was rising with a clear eye on the typhoon, was caught in the Jeju-Seo Sea, which is greatly affected by the 20 typhoon Shimaron. Solic moved on the edge of the North Pacific meat. As the Cimarron came in between, the meat I pushed it to the east. The wind blowing from the Cimarron here pulled Solic to the south and slowed down the speed of the north. This is how the speed of a person's walk It also drops, and Solic stays in Jeju Seohae for 24 hours. Meanwhile, the storm of Jeju poured energy and water. I spilled it. I stayed in one place for a long time, and the cold water below the sea level was brought up, and I lost my strength suddenly. If you go up early here, you won't meet it. The strong wind of the storm came down and broke the structure that was warning the typhoon. Eventually, the eyes of the typhoon disappeared at this time and the shape of the typhoon was completely distorted. This weakened Solic is in the southern part of the Korean Peninsula As it landed in Mokpo, it was no longer able to supply water and lost its power and escaped the Korean Peninsula. SBS Eom Min-jae."
} |
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",
"H... | [
"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",
"Floo... | [
"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 f... | {
"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"
} | 32 | [
"A severe storm is affecting the coastline in Japan, with powerful waves crashing against the shore. The image shows rough sea conditions, likely due to the storm.",
"The current frame depicts powerful waves crashing against rocky outcrops in Japan's southern region, reflecting the ongoing severe storm. The sea c... | Aiq8I97bscw | A severe tropical storm, known as Typhoon Solik, is affecting the coastline of Japan and moving northeastward towards Korea. The storm, which originated near Japan, has been tracked since August 16th and is expected to bring strong winds and heavy rain to the region. As of the 20th, the storm is positioned 850 km south of Tokyo and is forecasted to pass 270 km north of Jeju Island by the morning of the 22nd. The storm's trajectory and potential impact on different areas are being closely monitored, with warnings issued for heavy rain, strong winds, and high waves. The typhoon is expected to affect areas including Jeju Island, with waves potentially reaching up to 4 meters high. Residents are advised to exercise caution, prepare thoroughly, and be aware of the possibility of secondary disasters due to the high waves and strong winds. The storm's approach is compared to a significant typhoon that occurred in September 2012, emphasizing the potential severity of the situation. The video footage shows the storm's impact on coastal areas, with powerful waves crashing against rocky shores, flooded streets, and strong winds causing damage. Overall, the video provides updates and warnings about Typhoon Solik's trajectory, expected impact, and necessary precautions for residents in the affected areas. | {
"original": " Thank you. Thanks for watching!",
"refined": "Thank you. Thanks for watching!",
"summarized": "Not Available",
"translated": "Thank you for watching!",
"translated_llm": "Thank you for watching!",
"translated_whisper": " Hello! I'm going to show you how to make a beautiful and easy-to-use water-based drink. The water-based drink is a drink made by a professional water-based drinker. This drink is made by a professional water-based drinker. This drink is made by Thanks for watching!"
} |
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",
"H... | [
"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",
"Floo... | [
"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 f... | {
"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"
} | 32 | [
"A bus is driving through a flooded area, likely due to heavy rainfall, as depicted in the image. The bus is partially submerged in muddy water, with the surrounding environment suggesting a flood situation. The text on the image emphasizes the severity of the flooding, indicating that the area is experiencing sign... | 25LVZnXdx0A | The video depicts a severe weather event, specifically a heavy rainfall situation in the central and northern regions, where a "water bomb" of 374.5mm was reported overnight. The footage shows various scenes of flooding, including buses and cars driving through flooded streets, flooded indoor areas, and a parking garage with partially submerged vehicles. News reporters from YTN are shown providing updates on the situation, discussing the extent of the damage and the potential for further flooding. The video also includes scenes of emergency vehicles responding to the situation and people navigating through flooded areas. The reporters mention that the central region, which had already suffered damage from Typhoon Solly, is now experiencing even more severe damage from the current flooding. The video concludes with a message about being careful and mindful of the situation, both in terms of the weather and one's actions. Overall, the video highlights the destructive force of the heavy rainfall and the importance of caution and preparedness in the face of severe weather events. | {
"original": " 374.5mm의 물폭탄이 밤사이 중북부 지방에는 그야말로 물폭탄이 쏟아졌습니다. 그제부터 출입이 통제됐는데요. 시간당 30mm 안팎의 강원 북부 지역에 호우경 국지성 호우가 쏟아지고 있습니다. 강한 바람이 걱정이었는데요. 다행히 다행히 다행히 현재 바람이 거의 불지 않아 우려가 됐던 강풍에 따른 피해는 크게 없을 것으로 예상이 되는 상황입니다. 누구인가 지금 누가 기침 소리를 내었어 이미 많은 비로 집안이 약해진 상태에서 오늘 오후부터 밤사이 또다시 또다시 폭우가 예상되고 있습니다. 으아! 이 타임 이 타임 이 타임",
"refined": "A water bomb of 374.5mm was poured in the central and northern regions overnight. Since then, entry has been controlled. Heavy rain, with about 30mm per hour, is pouring in the northern part of Gangwon-do. I was worried about strong winds, but fortunately, the wind is almost not blowing, so it's unlikely there will be significant damage from strong winds. With the house already weakened due to heavy rain, another storm is expected to hit between this afternoon and tonight.",
"summarized": "A water bomb with 374.5mm of rainfall hit the central and northern regions overnight, causing entry restrictions. Although strong winds were a concern, fortunately, they have been minimal. However, with homes already weakened by previous heavy rain, another storm is expected in the afternoon and evening, bringing further rainfall to the northern area of Gangwon-do.",
"translated": "A 374.5mm water bomb was poured down overnight in the central and northern regions. Since then, access has been restricted. In the northern part of Gangwon Province, localized heavy rain is pouring down at a rate of around 30mm per hour. There were concerns about strong winds, but fortunately, the winds are currently very weak, so it's expected that the damage from strong winds will be minimal. Who coughed? With homes already weakened from the heavy rain, another heavy downpour is expected from this afternoon to tonight.",
"translated_llm": "374.5mm water bombs in the middle of the night in the northern province almost poured water bombs. Since then the entry has been controlled. 30mm per hour in the northern area of the strong An of the northern region is pouring tropical rain. Strong winds were worried.",
"translated_whisper": " 374.5mm water bomb Water bomb was poured in the central and northern regions overnight. Since then, the entry has been controlled. 30mm of anteroom per hour The northern air is pouring in the northern area of Gangwon-do. I was worried about this strong wind. Fortunately, fortunately, fortunately, the wind is almost not blowing. It is expected that there will be no damage due to the strong wind that was a concern. Who is it? Who is coughing now? With the house already weak due to heavy rain, another storm is expected between today afternoon and night. Bye!"
} |
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",
"H... | [
"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",
"Floo... | [
"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 f... | {
"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"
} | 32 | [
"The frame shows significant structural damage from a collapsed construction site, with scaffolding and debris scattered around. The church steeple and hotel roof are visibly affected by strong winds, suggesting the aftermath of severe weather. Large trees are uprooted, indicating the destructive power of the winds... | BRxOTPDR0g4 | The video depicts the aftermath of a severe weather event, likely caused by Typhoon Solik, which is compared to Typhoon Gompas from 2010 in terms of its destructive power. The scenes show significant structural damage, including a collapsed construction site, uprooted trees, and debris scattered across streets and sidewalks. The typhoon's impact on urban areas is highlighted, with damaged buildings, affected public facilities, and disrupted daily life.
The video also illustrates the typhoon's effects on natural environments, such as beaches and agricultural areas, with scenes of turbulent seas, fallen trees, and damaged greenhouses. The comparison to Typhoon Gompas, which caused 6 deaths, 11 injuries, and significant property damage, underscores the potential severity of Typhoon Solik's impact.
Throughout the video, maps and forecasts are shown, tracking the path of Typhoon Solik as it moves towards the Korean Peninsula, with predictions of heavy rainfall and strong winds. The scenes of people struggling with umbrellas, reduced visibility due to heavy rain and mist, and swaying palm trees emphasize the challenging weather conditions.
The video concludes with a focus on the severe weather conditions caused by Typhoon Solik, including heavy rainfall and strong winds, with text overlays providing updates on the typhoon's movement and expected impact. The final frames feature a Venn diagram representing the intersection of high wind speeds and heavy rainfall, highlighting the multifaceted nature of the typhoon's impact. | {
"original": " 건물 공사 구조물이 무너지고 교회 첨탁과 호텔 지붕도 강풍에 날려 추락했습니다. 아름드리 가로수는 뿌리째 뽑혔습니다. 지난 2010년 태풍 곰파수가 물고 온 강풍의 위력입니다. 전남 홍도에 초속 42.9m의 폭풍이 불었고 태안 38.1m, 군산 33.6m, 서울도 초속 21.6m의 강풍이 기록됐습니다. 바람 태풍 곰파스로 6명이 숨지고 11명이 부상했으며 1,673억 원의 재산 피해가 발생했습니다 북상하는 19호 태풍 솔릭은 8년 전 곰파스와 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 곰파스보다 강한 세력을 유지한 채 북상할 가능성이 큽니다 기상청은 태풍 솔릭으로 인해 목요일과 금요일 사이 전국에 최고 400mm의 폭우가 쏟아지고 초속 40m가 넘는 강풍이 불 것으로 전망했습니다.",
"refined": "The building construction collapsed, and the church steeple and hotel roof were blown away by strong winds. The strong winds brought by Typhoon Gompas in 2010 were of the same magnitude. Jeollanam-do reported a storm with winds of 42.9m per second, Taean 38.1m per second, Gunsan 33.6m per second, and Seoul 21.6m per second. Six people died, 11 were injured, and property damage of 16.73 billion won occurred due to Typhoon Gompas. Typhoon Solik, moving north, is following a similar path to Gompas eight years ago. It is expected to pass west of Jeju Island tomorrow morning, then move north along the west coast, and make landfall on the Taean Peninsula, passing through the capital area. With sea temperatures around the Korean Peninsula reaching 28 degrees Celsius, there is a high likelihood that Solik will maintain its strong force as it moves north. The weather service forecasts that Typhoon Solik will bring heavy rainfall of up to 400mm nationwide between Thursday and Friday, with winds exceeding 40m per second.",
"summarized": "A typhoon similar to Gompas in 2010 is expected to hit the Korean Peninsula, with Typhoon Solik moving north along a similar path. Strong winds of over 40m per hour and heavy rainfall of up to 400mm are forecasted nationwide between Thursday and Friday. The storm is expected to pass the west coast of Jeju Island, land on the Taean Peninsula, and then move through the capital region. The high sea temperature of around 28 degrees Celsius may contribute to the storm maintaining its strength. The previous typhoon Gompas caused 6 deaths, 11 injuries, and 16.73 billion won in property damage.",
"translated": "The building construction collapsed, and the church steeple and hotel roof were blown away by the strong winds. The large trees were uprooted. This is reminiscent of the powerful winds brought by Typhoon Gompas in 2010. At that time, a storm with winds of 42.9m/s was recorded in Jeonnam's Hongdo, 38.1m/s in Taean, 33.6m/s in Gunsan, and 21.6m/s in Seoul. The typhoon Gompas resulted in 6 deaths, 11 injuries, and 1.673 trillion won in property damage. The 19th typhoon, Solik, is moving along a similar path to Gompas eight years ago. It is expected to pass west of Jeju Island tomorrow morning, move up the west coast, and make landfall on the Taean Peninsula, passing through the metropolitan area. As the sea temperature around the Korean Peninsula is around 28 degrees Celsius, which is relatively high, there is a high possibility that Solik will maintain its strong force as it moves north. The weather agency forecasts that due to Typhoon Solik, heavy rainfall of up to 400mm will fall nationwide between Thursday and Friday, and strong winds of over 40m/s will blow.",
"translated_llm": "The storm hit the entire province of South Korea at a speed of 42.9m/s, and the storm hit the entire province of South Korea at 38.1m/s, Gunsan at 33.6m/s, and Seoul at 21.6m/s. The storm killed 6 people, injured 11 people and caused property damage of 1,673 billion won. The 19th typhoon Solrik is moving in a similar path to Bear Pass eight years ago.",
"translated_whisper": " Construction of the building collapsed and the church ceiling and hotel roof were blown away by the strong wind. The Arumdeuri River was the third largest in the country. The strong wind has been the cause of the typhoon Gompas in 2010. A storm of 42.9 meters in the early hours of the day in the south of Jeonnam was blown, and a strong wind of 33.6 meters in the day in the south of Taean, and a strong wind of 21.6 meters in the early hours of the day in Seoul were recorded. 6 people died and 11 people were injured due to the wind storm Gompas. The damage to the property of 16.73 billion won was caused. The storm Solik, which is in the north, is moving in a similar direction to the Gompas 8 years ago. Tomorrow morning, it is expected to pass the west coast of Jeju Island and then land on the Taean Peninsula and pass the capital. The temperature around the Korean Peninsula is 28 degrees Celsius, which is very high, and it is likely to be in the north with a stronger force than the Gompas. The weather service predicted a storm of 400mm across the country between Thursday and Friday due to typhoon Solik and a strong wind of over 40m per hour."
} |
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,... | [
"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 ... | [
"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... | {
"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"
} | 32 | [
"The frame shows a series of international flags in the background with a graphic of a gavel and the Korean text \"세상에 이런 법이\" (There's Such a Law in the World). The ASR indicates the video is discussing a massive sandstorm in Dunhuang, China, describing its impact on visibility and its unusual occurrence in summer... | Yddr-IKR_KA | A massive sandstorm, described as a yellow dust storm, engulfed the city of Dunhuang in western China, reducing visibility on the highway to just 5 meters ahead for drivers. The sandstorm, reportedly 100 meters high and several kilometers long, is unusual for the summer season, as such events typically occur between March and May. The storm forced drivers to turn on their headlights during the day due to poor visibility, making it hazardous to navigate the roads. A news anchor reported on the event, emphasizing its severity and the trend of increased frequency of large-scale sandstorms in northern and northwestern China. The video footage showed the sandstorm's impact on the city, with buildings and highways shrouded in a thick, dusty haze, and drivers struggling to see ahead. The anchor noted that this type of sandstorm is rare in the summer, but its occurrence is part of a larger trend of more frequent sandstorms in the region. | {
"original": " 지금까지 생생지구촌이었습니다. 눈에 보이지 않습니다. 정체는 다름 아닌 모래폭풍이라고 합니다. 와 저렇게 몰려오니까요. 고속도로라고 하는데 마치 연기 덩어리처럼 보이는 거대한 잿빛 모래폭풍이 순식간에 도로를 집어삼킵니다. 저게 높이가 100미터나 된다고요? 높이가 100미터뿐만 아니라 길이가 무려 수 킬로미터 정도에 달할 정도로 거대한 모래폭풍이었다고 합니다. 이 정체가 황사의 일종이었는데 지난 25일 중국 서부 간수성 둔함 도심을 완전히 저렇게 집어삼켜버렸어요. 먼지 구름을 동반한 모래폭풍이 순식간에 고속도로를 덮치게 되면서 당시 도로에 있던 운전자들이 5m 앞 시야도 확보가 되지 않았다는 거예요. 드랩박스 보니까 진짜 5m 앞에 안 보여요. 그렇죠. 대낮에도 지금 거의 밤처럼 밤도 저렇게는 시야 확보가 저 정도까지는 아닐 것 같은데 전조등을 켜고 운전을 해야 될 정도로 앞이 보이지 않았다고 합니다. 보통 황사가 3월에서 5월 정도 사이에 나타나는 여름 황사는 굉장히 드물잖아요. 그런데 지금 최근 중국 북부와 서북부 지역에서 이런 대규모 모래폭풍이 잇따라 발생을 하고 있다고 합니다.",
"refined": "A severe sandstorm has occurred in China. It's massive, covering the entire city in darkness in an instant. On the highway, a gigantic, 100-meter-high sandstorm that resembles a lump of smoke swallowed the road in a matter of seconds. Not only was it 100 meters high, but it also stretched for several kilometers. This massive sandstorm, a type of yellow sand, completely engulfed the city of Dunhuang in western China's Gansu province on the 25th. The sandstorm, accompanied by dust clouds, quickly covered the highway, reducing visibility for drivers to just 5 meters ahead. Looking at the dashcam footage, it's clear that visibility was severely impaired. Even in broad daylight, it was like driving at night, with drivers forced to use their headlights due to the poor visibility. Typically, yellow sandstorms are rare between March and May, but recently, large-scale sandstorms have been occurring in northern and northwestern China.",
"summarized": "A massive sandstorm, approximately 100 meters high and several kilometers long, swept through the city of Dunhuang in western China on the 25th. The storm, a type of desert sandstorm, reduced visibility to just 5 meters, forcing drivers to use their headlights in the daytime. This type of sandstorm is rare in the summer months, but similar large-scale sandstorms have recently occurred in northern and northwestern China.",
"translated": "That's all for today's Live Earth Village. You can't see it, but it's actually a massive sandstorm. The highway, which is normally bustling, is suddenly engulfed in a huge, gray sandstorm that looks like a cloud of smoke, sweeping across the road in an instant. It's reportedly 100 meters high, and not only that, but it's also several kilometers long. This massive sandstorm, a type of yellow dust storm, completely engulfed the city of Dunhuang in western China on the 25th. The sandstorm, accompanied by a dust cloud, suddenly covered the highway, reducing visibility to just 5 meters ahead for drivers on the road at the time. As you can see from the dashcam footage, it's almost impossible to see 5 meters ahead. It's like driving at night, even in the daytime, with visibility so poor that drivers had to turn on their headlights. Typically, yellow dust storms occur between March and May, but it's rare to see such a massive sandstorm in the summer. However, recently, large-scale sandstorms like this have been occurring frequently in northern and northwestern China.",
"translated_llm": "It was a massive sandstorm, not only 100 meters high but also many kilometers long. It was a typhoon that had been raging for 25 days and completely engulfed the entire western Chinese city. The duststorm, accompanied by a yellow sandstorm, had suddenly covered the highway, and the drivers on the road at that time were not able to see the road. The drabstorm was not really visible until 5 am.",
"translated_whisper": " The world is telling you about various cases in various countries. In China, a super-large sandstorm is coming. Wow, you've never seen that much, have you? The whole city is covered in darkness in a blink of an eye. Let's look at the screen. Just like a bomb, The huge amount of light and smoke spreads rapidly, and you can't see the apartment complex for a moment. It's a sandstorm. Wow, it's so crowded. It's called a highway, and it's like a huge sandstorm that looks like a lump of smoke swallows the road in an instant. That's 100 meters high? It was a huge sandstorm that was not only 100 meters high but also several kilometers long. It was a kind of desert, but on the 25th, it completely swallowed the city of Dunhuang, a suburb of the western part of China. The sandstorm accompanied by dust and clouds covered the highway in an instant, and the drivers on the road at the time couldn't even see 5 meters ahead. I can't see 5 meters ahead when I look at the black box. Right. It's almost night time now. It's not that late at night, but it's not that far away that you have to turn on the light and drive. Usually, there are very few summer phoenixes between March and May. Recently, a large-scale sandstorm like this has been occurring in the north and southwestern regions of 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,... | [
"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 ... | [
"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... | {
"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"
} | 32 | [
"A weather anchor in a news studio is reporting on a massive sandstorm that originated in Mongolia, passed through China, and reached the Korean Peninsula. The backdrop of the studio shows footage of the sandstorm, highlighting its impact as it engulfs areas and creates a thick haze. The text on the screen mentions... | 3BbP2OHdePY | A massive sandstorm that originated in Mongolia passed through China and reached the Korean Peninsula, causing severe air pollution and reducing visibility. The storm is expected to mix with rain, resulting in a muddy downpour across the country, with rain starting in the central region in the morning and spreading nationwide by night. Strong winds, thunder, and lightning are forecasted for the afternoon, with possible hail in some regions. The weather agency has warned of potential damage to facilities and crops and advised caution. The sandstorm has already caused significant environmental impact, with high concentrations of fine dust in western coastal regions, and the upcoming muddy rain is expected to exacerbate the situation. Citizens are taking precautions, wearing masks and cleaning their vehicles to cope with the dusty conditions. The weather report emphasizes the need for preparedness and vigilance to mitigate the effects of the severe weather conditions. | {
"original": " 뿌연 하루였습니다. 몽골에서 시작된 모래폭풍이 중국을 거쳐서 한반도에 상륙한 건데 내일은 전국에서 황사 가루가 섞인 흙빛까지 내릴 전망입니다. 최원영 기자입니다. 거대한 모래폭풍이 마을을 통째로 집어삼켰습니다. 사흘 전 중국 북서부 간수성의 한 마을에서 촬영된 영상입니다. 지난 주말 몽골에서 만들어진 이 모래폭풍이 중국을 휩쓴 뒤 오늘부터 국내로 유입되기 시작했습니다 항공기에서 내려다본 아래가 온통 뿌였습니다 도심도 황사스모그에 휩싸였습니다 주차된 차에서는 먼지가 묻어나옵니다 오늘 새벽 백령도에 상륙한 황사는 북서풍을 타고 수도권과 서해안으로 점차 확대됐습니다 미세먼지 농도는 광주, 대구, 천안 등 서해안을 중심으로 나쁨 수준을 보였습니다. 내일은 황사 때문에 흙빛까지 내립니다. 비는 오전 중부지방부터 내리기 시작해 밤에는 전국으로 확대됩니다. 북쪽으로부터 남한 찬 공기로 인해 대기가 매우 불안정해지면서 전국 전국 곳곳에 비가 내리겠습니다. 오후에는 돌풍과 함께 천둥, 번개가 치는 곳도 있겠습니다. 기상청은 강원도와 충북 북부, 경북, 경남 일부 지역에서는 우박이 떨어질 수 있으니 시설물과 농작물 피해에 주의해달라고 당부했습니다. TV조선 최한영입니다. 감사합니다.",
"refined": "It was a cloudy day. The sandstorm that started in Mongolia has passed through China and landed on the Korean Peninsula. Tomorrow, it is expected to rain with yellow dust mixed with black dust throughout the country. This is Choi Won-young. A huge sandstorm engulfed an entire village. This video was filmed in a village in northwest China's Gansu Province four days ago. The sandstorm that formed in Mongolia last weekend swept through China and began to affect Korea today. The ground below was completely covered in dust when viewed from an airplane. The city was also shrouded in a haze. Dust covered parked cars. The yellow dust that landed on Baekryeong Island this morning gradually spread to the metropolitan and western coastal areas via a northwest wind. The fine dust concentration was particularly high in the west coast, including Gwangju, Daegu, and Cheonan, reaching bad levels. Tomorrow, the dust is expected to mix with rain. Rain will start in the central region in the morning and expand nationwide by night. Due to cold air from the north, the atmosphere is very unstable, leading to rain in various parts of the country. In the afternoon, some areas may experience thunderstorms. The weather office warned that hail may fall in some areas, including Gangwon-do, northern Chungcheongbuk-do, and parts of Gyeongsangbuk-do and Gyeongsangnam-do, and advised caution to prevent damage to facilities and crops.",
"summarized": "A massive sandstorm originating in Mongolia has passed through China and reached the Korean Peninsula, bringing with it a mix of yellow dust and black dust. The storm is expected to continue, with rain and strong winds forecasted for the entire country tomorrow. Fine dust concentrations are already at bad levels in western coastal areas such as Gwangju, Daegu, and Cheonan. The Korea Meteorological Administration has warned of potential damage to facilities and crops, particularly in Gangwon-do, Chungbuk, Gyeongbuk, and Gyeongnam, where hail is possible.",
"translated": "It was a hazy day. A massive sandstorm that originated in Mongolia passed through China and has landed on the Korean Peninsula. Tomorrow, it's expected to rain, with yellow dust and fine dust particles mixing to create a muddy downpour across the country. I'm Choi Won-young. A gigantic sandstorm engulfed an entire village. This footage was taken in a village in Gansu Province, northwest China, four days ago. The sandstorm, which formed in Mongolia over the weekend, swept through China and began to affect Korea today. As seen from an airplane, the ground below was completely obscured by the dust. The city was also shrouded in a thick haze. Even parked cars were coated in a layer of dust. The yellow dust that landed on Baekryeong Island this morning spread to the metropolitan area and the west coast, carried by northwest winds. The concentration of fine dust was particularly high in western coastal regions, including Gwangju, Daegu, and Cheonan, reaching unhealthy levels. Tomorrow, the dust is expected to mix with rain, resulting in a muddy downpour. Rain will start falling in the central region in the morning and spread nationwide by night, as cold air from the north destabilizes the atmosphere. In the afternoon, some areas can expect strong gusts of wind, thunder, and lightning. The weather agency has warned of possible hail in some regions, including Gangwon-do, northern Chungcheongbuk-do, and parts of Gyeongsangbuk-do and Gyeongsangnam-do, and has advised caution to prevent damage to facilities and crops. I'm Choi Han-young, reporting for TV Chosun. Thank you.",
"translated_llm": "This is a video shot in a village in northwest China. This is a video shot in a village in northwest China. This is a video shot in the past weekend. This sandstorm originated in Mongolia and swept through China and began to be introduced into the country from today. It came down from the plane and hit the city center. It hit the city center.",
"translated_whisper": " It was a cloudy day. The sandstorm that started in Mongolia has passed through China and landed on the Korean Peninsula. Tomorrow, it is expected to rain even black dust mixed with yellow dust throughout the country. This is Choi Won-young. A huge sandstorm swallowed the whole village. This video was filmed in a village in Gansu-seong, northwest China, four days ago. This sandstorm created last weekend in Mongolia swept through China and has begun to enter Korea from today. The ground that was lowered from the aircraft was completely covered. The city was also covered in sulfur smoke. Dust comes out of parked cars. The sulfur that landed in Baekryeong-do this morning gradually expanded to the metropolitan and western coasts by a north-west wind. The fine dust concentration was in the west coast, such as Gwangju, Daegu, and Cheonan, and it was seen at a bad level. Tomorrow, the dust will even rain. Rain starts to fall in the central and central regions in the morning and expands to the whole country at night. Due to the cold air from the north, the atmosphere is very unstable and it will rain in the It will rain in the north of the country. In the afternoon, there will be a thunder and lightning There will be. The weather office in Gangwon-do and the north-south, In some areas of Gyeongbuk and Gyeongnam, I asked you to be careful of facilities and crops I asked. This is TV Choe Han-haeng."
} |
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,... | [
"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 ... | [
"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... | {
"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"
} | 32 | [
"A massive dust storm is approaching a city, with a large, dense cloud of dust obscuring the sky. The buildings in the foreground are overshadowed by the storm, suggesting the imminent impact of the dust storm on the urban area. The atmosphere appears ominous and foreboding.",
"The dust storm continues to advance... | 6sLn3wHQvCQ | A massive dust storm swept through a city in Gansu Province, Guazhou, China on July 25, engulfing the urban area in a dense cloud of dust. The storm, described as "dark and massive," reduced visibility to near zero, causing a significant threat to the residents. The dust storm was so intense that it was likened to a tsunami, with the sky turning completely orange due to the overwhelming amount of dust. The air quality was reported to be the worst, and a yellow alert was issued for the sandstorm. The storm caused a landslide in a mountainous area, which sent a large cloud of dust and debris cascading down the hillside, almost covering travelers below. The landslide continued to displace significant amounts of earth and debris, with people evacuating the area and vehicles being damaged. Fortunately, there were no reported casualties. The disaster was extensively covered by Yonhap News, with footage showing the devastating impact of the dust storm and landslide on the urban and natural environments. The video ended with a call to action to subscribe to Yonhap News. | {
"original": " Teksting av Nicolai Winther",
"refined": "Text by Nicolai Winther",
"summarized": "Not Available",
"translated": "Text by Nicolai Winther",
"translated_llm": "Text by Nicolai Winther",
"translated_whisper": ""
} |
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,... | [
"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 ... | [
"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... | {
"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"
} | 32 | [
"The current frame depicts a scene of intense atmospheric disturbance, likely a severe sandstorm, which is consistent with the original ASR transcript describing a massive sandstorm in Mongolia. The image shows a thick cloud of dust or sand obscuring the sky, with the visibility significantly reduced due to the swi... | qaan1wMGnN0 | A severe sandstorm originating from the dry grasslands of Mongolia engulfed the sky, threatening to swallow small villages and causing widespread disruption. The storm, accompanied by strong winds, reduced visibility and made it difficult to see ahead, affecting both the grasslands and surrounding cities. Approximately 600 Mongolian nomads were temporarily reported missing, but most were found alive, although around 10 people, including a 5-year-old child, lost their lives. The storm was powerful enough to topple a 10-story tower and caused power outages, leaving residents in darkness and cold. Rescue efforts were undertaken, including the rescue of a 71-year-old grandmother after police cleared snow and sand from her home. The storm's severity was exacerbated by the lack of spring grass in the Mongolian plains, resulting in an abundance of loose dust. Although the immediate storm subsided, forecasts indicate that several more yellow sandstorms are expected this spring, posing an ongoing threat to the region. | {
"original": " 거대한 모래폭풍이 하늘을 뒤덮었습니다. 지난 주말 몽골의 건조한 초원에서 발원한 황사바람입니다. 태풍급 강풍을 동반한 모래먼지가 한적한 작은 마을을 집어삼킬 듯한 기세입니다. 초원은 물론이고 주변 도시들도 앞을 분간하기 어려운 먼지 속에 갇혔습니다. 황색폭풍의 습격으로 몽골 초원지대 유목민 약 600명이 한때 실종됐습니다. 다행히 대부분 살아서 돌아왔지만 5살 어린아이를 포함해 10여 명은 결국 숨졌습니다. 최고 풍속 30미터가 넘는 모래폭풍에 10층 높이의 철탑도 맥없이 부러졌습니다. 인근 지역 주민들은 모래바람의 공포 속에 정전까지 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 눈모래를 걷어낸 지 두 시간 만에 71살 할머니가 구조되기도 했습니다. 이번 황사는 강한 바람이 모래먼지를 멀리까지 실어 나르면서 10년 만에 황사로 강해졌습니다. 특히 몽골 초원에 봄풀이 아직 자라기 전이어서 흙먼지가 많았다고 중국기상당국은 분석했습니다. 황사로 뒤덮였던 베이징에는 다시 맑은 날씨가 찾아왔습니다. 강한 바람 때문에 모래먼지가 하루 만에 사라진 건데 이번 봄에 몇 차례 더 황사가 예보가 돼 있는 상황입니다.",
"refined": "A massive sandstorm engulfed the sky, spawned from Mongolia's arid grasslands last weekend. The sandstorm, accompanied by typhoon-force winds, threatened to engulf a small, quiet village. The grasslands and surrounding cities were shrouded in dense dust, making it difficult to distinguish anything. The yellow storm left approximately 600 Mongolian nomads missing, but fortunately, most were later found alive. However, 10 people, including a 5-year-old child, ultimately lost their lives. The intense storm, with winds exceeding 30 meters per second, was strong enough to topple a 10-story steel tower. Nearby residents, gripped by fear of the sandstorm, were plunged into darkness and cold due to a power outage. In Naimonggu, northwestern China, near the Mongolian grasslands, a sandstorm mixed with snow swept through the area. Strong winds drove sand and snow into homes, trapping residents inside. In one instance, police used shovels and carts to clear the snow and sand, rescuing a 71-year-old grandmother after two hours. This yellow sandstorm was particularly severe, with strong winds dispersing sand dust over a wide area, making it the strongest in a decade. According to Chinese weather authorities, the severity was exacerbated by the fact that the spring grass in Mongolia had yet to grow, resulting in an abundance of loose soil. Clear weather has returned to Beijing, which was previously blanketed with yellow dust. The sand dust dissipated within a day due to strong winds, but several more yellow sandstorms are forecast for this spring.",
"summarized": "A massive sandstorm swept across Mongolia's dry grasslands and surrounding areas in China, causing widespread disruption and loss of life. The storm, which brought typhoon-like winds and reduced visibility, left around 600 nomads temporarily missing, with 10 people, including a 5-year-old child, eventually succumbing to the extreme weather. The strong winds, reaching speeds of over 30 meters per second, also caused a 10-story tower to collapse. In neighboring China, the storm brought snowstorms mixed with sand, trapping residents in their homes and requiring police rescue efforts. Authorities attributed the storm's severity to the strong winds and the lack of spring growth in the Mongolian grasslands, which contributed to the large amount of dust in the air. Although the weather has cleared in Beijing, the region is expected to experience several more sandstorms this spring.",
"translated": "A massive sandstorm engulfed the sky, originating from the dry grasslands of Mongolia over the weekend. The storm, accompanied by typhoon-like winds, threatened to swallow a small, tranquil village. Both the grasslands and surrounding cities were shrouded in a thick dust that made it difficult to see ahead. Approximately 600 Mongolian nomads were temporarily reported missing due to the yellow storm, but fortunately, most were found alive. However, the storm ultimately claimed the lives of around 10 people, including a 5-year-old child. The intense sandstorm, with gusts exceeding 30 meters per second, was so powerful that it even toppled a 10-story tower. Residents in nearby areas lived in fear of the sandstorm, experiencing power outages that left them in darkness and cold. In Naimonggu, located in northern China near the Mongolian plains, a mixture of sand and snowstorms hit the area. Strong winds blew sand and snow into homes, trapping residents inside. In one instance, a 71-year-old grandmother was rescued after police spent two hours clearing the snow and sand with shovels and carts. This yellow sandstorm was particularly severe due to strong winds carrying dust over long distances, a phenomenon that has not been seen in 10 years. According to Chinese weather authorities, the storm's severity was exacerbated by the fact that spring grass had not yet grown in the Mongolian plains, resulting in an abundance of loose dust. Although Beijing, which was previously covered in yellow dust, has since seen clearer skies, the strong winds caused the sand dust to dissipate within a day. Nonetheless, forecasts indicate that several more yellow sandstorms are expected this spring.",
"translated_llm": "A huge sandstorm covered the sky. Last weekend, a yellow wind blew up in the Mongolian dry grasslands. The winds of a typhoon and a strong wind seemed to engulf a small village. The grasslands, of course, and the surrounding cities were trapped in the hard-to-break dust. The storm caused the death of about 600 people in the Mongolian grasslands.",
"translated_whisper": " A huge sandstorm covered the sky. It is a yellowish wind that erupted from the dry grasslands of Mongolia last weekend. The sand dust accompanied by a typhoon-like strong wind is like swallowing a quiet small village. The grasslands and surrounding cities were also trapped in the dust that was difficult to distinguish. About 600 Mongolian nomads were once missing due to the yellowish storm. Fortunately, most of them came back alive, but 10 people, including 5-year-old children, eventually died. A sandstorm over 30 meters high and a 10-story tower was broken. The residents of the nearby area were in the fear of the sandstorm, and they fell into darkness and cold. In Naimonggu, northwest China, close to the Mongolian plain, snowstorms mixed with sand were scattered. Sand and snow that had been pushed by the strong wind covered the houses and the residents who were trapped in the house were also rescued. Two hours after the police took out the snow and the sand with a shovel and a cart, a 71-year-old grandmother was rescued. This yellow sand was strong as the strong wind blew away the sand dust and became strong as a yellow sand in 10 years. In particular, the Chinese weather authorities analyzed that there was a lot of black dust because the spring grass in the Mongolia was not yet growing. The clear weather came back to Beijing, which was covered with yellow dust. Sand dust disappeared in a day because of strong winds, but this spring, yellow dust has been reported several times."
} |
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 ... | [
"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, a... | [
"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... | {
"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"
} | 32 | [
"The current frame shows a news broadcast with a weather update. A weather map is displayed on the left side, indicating areas of high heat and possibly heat wave warnings across South Korea. The map shows regions colored in red and orange, signifying high temperatures. The text in Korean indicates the types of hea... | 0ZROpcl5nUE | A female newscaster presents a weather forecast on a sunny beach setting, discussing an ongoing heat wave in South Korea. The heat wave warning has been expanded and strengthened, with the highest temperatures recorded in Gyeongju, Daegu, and Seoul. Many areas will experience little relief from the heat at night, leading to tropical nights. The forecast shows that tomorrow, Seoul will reach 33 degrees and Daegu will reach 37 degrees during the day, with clear skies and strong UV rays expected. The heat wave is expected to continue, with no significant rain forecast, and the newscaster advises caution due to the high temperatures and strong UV rays. The forecast persists throughout the video, with the newscaster reiterating the need for continued caution against the heat and strong UV rays. | {
"original": " 열기가 식을 새도 없이 계속 쌓여만 가고 있습니다. 폭염특보도 확대, 강화되고 있는데요. 오늘 오전 11시 일기에 경기 남부와 강원 영서 남부에도 폭염경보가 내려졌고요. 전국 대부분 지역으로 폭염주의보가 확대됐습니다. 현재 경북 경주기온이 36.6도로 전국에서 가장 뜨겁고요. 대구 36.4도, 서울도 32.9도까지 올라서 올해 최고기온을 기록할 것으로 보입니다. 밤에도 열기는 쉽게 식지 못하고 열대야로 이어지는 곳이 많겠는데요. 내일 서울의 아침 기온은 24도, 대구 25도로 오늘 밤 사이에도 더위로 잠 못 이루는 곳이 많겠고요. 낮에는 다시 서울 기온이 33도, 대구는 37도까지 오르는 등 오늘과 비슷하겠습니다. 현재 전국에 맑은 하늘이 드러나 있는데요. 내일도 종일 쾌청한 하늘이 함께하겠지만 강한 자외선에는 주의하시는 게 좋겠습니다. 당분간 뚜렷한 비 소식 없이 낮에는 폭염이 밤에는 열대야가 기승을 부리겠습니다 지금까지 날씨정보였습니다",
"refined": "The heat continues to build up without cooling down. The heat wave warning has been expanded and strengthened. This morning at 11 a.m., a heat wave warning was issued for the southern part of Gyeonggi and the southern part of Yeongseo in Gangwon Province, and a heat wave advisory has been expanded to most areas of the country. The current temperature in Gyeongju, Gyeongsangbuk-do, is 36.6 degrees, the highest in the country. Daegu is 36.4 degrees, and Seoul is 32.9 degrees, which may record the highest temperature of the year. Many places will not cool down easily even at night, leading to tropical nights. Tomorrow morning, the temperature in Seoul will be 24 degrees, and in Daegu, it will be 25 degrees, so many places will have difficulty sleeping due to the heat. In the afternoon, the temperature in Seoul will rise to 33 degrees, and in Daegu, it will rise to 37 degrees, similar to today. Although the sky is currently clear across the country, it will remain clear all day tomorrow, but strong UV rays are a concern. For the time being, there will be no significant rain, and the heat wave will continue during the day, while tropical nights will persist at night.",
"summarized": "A heat wave warning has been issued for most areas of the country, with temperatures expected to remain high throughout the day and night. The highest temperature was recorded in Gyeongju, Gyeongsangbuk-do, at 36.6 degrees, while Seoul and Daegu reached 32.9 and 36.4 degrees, respectively. Overnight temperatures will remain high, with Seoul at 24 degrees and Daegu at 25 degrees, making it difficult to sleep. Tomorrow's forecast is similar, with temperatures rising to 33 degrees in Seoul and 37 degrees in Daegu. Clear skies are expected, but strong UV rays are a concern.",
"translated": "The heat continues to build up without letting up. The heat wave warning has also been expanded and strengthened. As of 11 am today, a heat wave warning was issued for the southern part of Gyeonggi and the southern part of Gangwon's Yeongseo region, and the heat wave alert has been expanded to most areas of the country. Currently, the temperature in Gyeongju, Gyeongsangbuk-do, is 36.6 degrees, the highest in the country, followed by Daegu at 36.4 degrees and Seoul at 32.9 degrees, which is expected to be the highest temperature of the year. Many places will experience little relief from the heat even at night, leading to tropical nights. Tomorrow, the morning temperature in Seoul is expected to be 24 degrees and 25 degrees in Daegu, with many areas likely to experience another sleepless night due to the heat. During the day, temperatures are expected to rise again, with Seoul reaching 33 degrees and Daegu reaching 37 degrees, similar to today. Currently, the sky is clear across the country, and it is expected to remain clear throughout the day tomorrow, but caution is advised due to strong UV rays. For the time being, there is no significant rain forecast, and the heat wave is expected to continue during the day, with tropical nights persisting. That's the latest weather update.",
"translated_llm": "The temperature is increasing without any warning. The temperature is increasing. The temperature in Seoul is increasing. At 11 am, the temperature in the southern and central regions of the country has been reduced. The temperature in the northern regions is increasing. Currently, the temperature in the northern regions is 36.6 degrees.",
"translated_whisper": " The heat is still piling up without a sign of cooling. The heat wave is also being increased. On Sunday morning 11 a.m., a heat wave warning was issued in the southern part of Gyeonggi and Yeongseo of Gangwon Province. The heat wave warning was expanded to most areas of the country. The current temperature in Gyeongju, Gyeongsuk, is 36.6 degrees, the hottest in the country. Daegu is 36.4 degrees, Seoul is 32.9 degrees, and it seems to record the highest temperature of the year. There are many places that are not easy to cool down and continue to heat up even at night. Tomorrow, Seoul's morning temperature is 24 degrees and Daegu is 25 degrees, so there are many places that cannot sleep due to the heat even during the night. In the afternoon, Seoul's temperature will rise again to 33 degrees and Daegu will rise to 37 degrees, which will be similar to today. The clear sky is now visible throughout the country. The sky will be clear all day tomorrow, but it is better to be careful of strong UV rays. There will be a strong wind and a strong rain in the afternoon. This is the weather report."
} |
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 ... | [
"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, a... | [
"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... | {
"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"
} | 32 | [
"In the current frame, we see a scene at a train station platform where several people are present. A person wearing a high-visibility vest stands near the train, holding a green object, possibly a tool or a bag. Another individual in blue attire is walking across the platform. The train is stopped at the platform,... | -HJV2kcDUAA | The video depicts a scene at Dongjak Station in South Korea, where a screen door malfunction has occurred, likely due to high temperatures. The incident is being addressed by individuals in high-visibility vests, who are inspecting and possibly repairing the issue. The train remains stationary at the platform with its doors open, and passengers are visible through the windows. The digital display board above the platform shows information in Korean, including the train schedule and warnings. As the video progresses, it shows the interior of the train, including an electrical control panel and a digital display screen announcing the upcoming War Memorial Day. The scene then shifts to a demonstration or simulation of an emergency procedure, where a person presses the emergency stop button and interacts with the control panel. The video concludes with scenes of the train in motion, passengers waiting at the platform, and the interior of the train with illuminated warning lights, indicating preparation for departure. Throughout the video, the text "연합뉴스" (Yonhap News) is visible, indicating the source of the footage. The overall context suggests a news report or documentary about the screen door malfunction and the safety procedures in place at Dongjak Station. | {
"original": " Thank you for watching. Have a nice day!",
"refined": "Thank you for watching. Have a nice day!",
"summarized": "The speaker bids farewell to the audience, thanking them for watching and wishing them a nice day.",
"translated": "Thank you for watching. Have a nice day!",
"translated_llm": "Thank you for watching.",
"translated_whisper": " Thank you for watching. Have a nice day!"
} |
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 ... | [
"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, a... | [
"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... | {
"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"
} | 32 | [
"The current frame shows a news anchor standing in a studio, with a backdrop image of a natural scene featuring a fenced-off area with large rocks and a waterfall. The image suggests a focus on an unusual weather event where ice pillars were observed in a hot climate. The text on the screen is in Korean and transla... | MZR5vPcBcI4 | The video reports on a unique natural phenomenon in Gyeongnam Miryang's Ice Valley, where fist-sized ice pillars have remained frozen despite the extreme heat. The ice valley, located on Jeyaksan Mountain, has a cool breeze blowing through the rocks, with an internal temperature of 9.2 degrees Celsius, similar to autumn weather. This natural wonder is a local landmark and designated as Natural Monument No. 224. The video shows visitors, exhausted from the heat, being amazed by the cool wind and ice pillars, with over 20,000 people visiting the site last month to seek refuge from the record-breaking heat. The ice typically melts during the rainy season, but this year it has persisted, making the ice valley a unique summer retreat for those weary of the heat. The reporter, Choi Jin-seok, documents the phenomenon, measuring the temperature inside the rocks and showcasing the ice formations, highlighting the contrast between the external heat and the cool interior of the valley. The video emphasizes the ice valley's popularity as a refreshing destination, offering a cool and natural escape from the extreme heat. | {
"original": " 극한 폭염 속에도 경남 밀양 얼음골에선 주먹만한 얼음기둥이 관측됐습니다. 신기한 건 이례적인 더위 속에도 이 얼음이 평년보다 더 오랫동안 꽁꽁 얼어있다는 건데요. 어떻게 된 일일까요? 최진석 기자가 소개합니다. 경남 밀양 제약산 해발 600m. 불볕 더위에 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 여기까지 올라오는데 참 많이 힘들었는데 여기 보니까 또 너무나 선선한 바람이 보니까 너무너무 좋아요 바위 틈에서는 에어컨을 틀어놓은 것처럼 차가운 자연 바람이 나오고 있습니다. 열화상 카메라로 온도를 재보니 1도밖에 나오지 않습니다. 이곳의 얼음은 보통 장마철인 6, 7원이면 다 녹아 없어지지만 올해는 계속되는 폭염의 상황에도 더 꽁꽁 얼어붙은 채 남아 있습니다. 기록적인 폭염 속에 더위를 식히려 지난달에만 2만여 명이 찾은 밀양 얼음골. 신비로운 비밀을 간직한 채 더위에 지친 이들의 색다른 피서지가 되고 있습니다. KBS 뉴스 최진석입니다.",
"refined": "In the midst of extreme heat, fist-sized ice pillars were observed in Gyeongnam Miryang's Ice Valley. Interestingly, despite the unprecedented heat, the ice has remained frozen for longer than usual. Reporter Choi Jin-seok has more on the story. In Gyeongnam Miryang's Jeyaksan mountain, 600 meters above sea level, the outside temperature has surpassed 36 degrees. However, a cool breeze blows through the rocks, and the temperature inside the rocks is a chilly 9.2 degrees. The area is home to a natural monument, Ice Valley, which features ice pillars in the rocks. Visitors who have been exhausted by the heat are amazed by the cool wind. Despite the difficulty of climbing up, the cool wind here is a welcome relief. The natural breeze through the rocks is as cold as an air conditioner, with temperatures measuring just 1 degree on a thermal imaging camera. Typically, the ice in this area melts away during the rainy season, but this year, it has remained frozen despite the prolonged heatwave. Miryang Ice Valley has become a unique summer destination for over 20,000 people who visited last month to escape the record-breaking heat, while keeping its mysterious secret intact. I'm Choi Jin-seok, reporting for KBS News.",
"summarized": "In the midst of extreme heat, an unusual phenomenon was observed in Gyeongnam Miryang's Ice Valley, where ice pillars were found to be frozen for a longer period than usual. The area, a natural monument, typically experiences temperatures around 9.2 degrees Celsius, despite the outside temperature being over 36 degrees Celsius. The ice in the valley usually melts during the rainy season, but this year it has remained frozen, attracting over 20,000 visitors last month seeking relief from the heat.",
"translated": "In the midst of extreme heat, fist-sized ice pillars were observed in Gyeongnam Miryang's Ice Valley. What's remarkable is that despite the unprecedented heat, the ice has remained frozen for longer than usual. Reporter Choi Jin-seok has more. Located in Gyeongnam Miryang, Jeyaksan Mountain has an elevation of 600 meters. Although the outside temperature has surpassed 36 degrees, a cool breeze blows through the rocks. The temperature inside the rocks measures a cool 9.2 degrees, similar to autumn weather, and even features fist-sized ice pillars. This natural wonder is a local landmark and designated as Natural Monument No. 224. Visitors, exhausted from the heat, are amazed by the cool wind. Despite the challenging climb, they are rewarded with a refreshing breeze. The air between the rocks is cool, much like the air from an air conditioner, with a temperature of only 1 degree, as measured by a thermal imaging camera. Typically, the ice melts away during the rainy season in June and July, but this year, it remains frozen despite the ongoing heatwave. Over 20,000 people visited Miryang Ice Valley last month, seeking refuge from the record-breaking heat. This enigmatic destination has become a unique summer retreat for those weary of the heat. I'm Choi Jin-seok, reporting for KBS News.",
"translated_llm": "The ice column is the 224th most famous natural monument in the region. The sun is shining on the ice. People who have seen the ice column have seen the ice column for a long time. It is very difficult to climb up here this year, but the weather is very difficult to find because the wind is usually too strong. The ice column is still very cold.",
"translated_whisper": " In the midst of extreme heat, ice pillars like fists were observed in Gyeongnam Miryang Ice Valley. Interestingly, in this unprecedented heat, the ice has been frozen for longer than usual. What happened? Reporter Choi Jin-seok will tell you. Gyeongnam Miryang mountain, Jeyaksan, has a sea level of 600m. The outside temperature of the hot hot hot water is over 36 degrees, but a cool breeze comes out through the rocks. When I measured the temperature inside the rock, it was 9.2 degrees in the cold autumn. There is even an ice pillar with a fist in the middle of the rock. It is the landmark of this area and the 224th ice ball of the Natural Monument. People who are tired of the hot weather are only amazed by the cool wind. It was very difficult to climb up here, but it's so nice to see the cool wind here. The rock team is getting cold natural wind like they turned on the air conditioner. The temperature is measured with a thermal imaging camera and it's only 1 degree. The ice here usually melts away in the rainy season, but this year, it's still frozen even in the situation of the continued storm. I've been observing it for decades, but it's been the longest. When it rains a lot in the rainy season, it goes down quickly, but it didn't rain a lot in the rainy season. Miryang Ice Bowl, which was found by more than 20,000 people last month to cool the heat of the record-breaking heat. It is a unique secret of those who are tired of the heat while keeping a mysterious secret. Choi Jin-seok, KBS News."
} |
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 ... | [
"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, a... | [
"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... | {
"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"
} | 32 | [
"The current frame shows a weather presenter standing in front of a weather forecast screen. The screen displays a graph indicating the temperature trends for Seoul over the next few days. The temperatures are shown in both red and orange lines, with red representing the daytime high and orange representing the nig... | HCV2Hx8B0-c | The video features a weather presenter providing an update on the ongoing heat wave in South Korea. The presenter stands in front of a screen displaying temperature trends for Seoul, indicating that the city has experienced its highest recorded temperature of 39.6 degrees Celsius. The forecast predicts that the heat wave will continue, with temperatures remaining high nationwide, especially in Seoul, where the temperature is expected to reach 39 degrees Celsius the following day. Although temperatures are expected to gradually decrease from Friday, the presenter emphasizes that it will still feel hot, with daytime temperatures around 35 degrees Celsius and morning temperatures above 25 degrees Celsius, leading to tropical nights in some areas. The presenter also mentions that the weather will be sunny nationwide, with a high UV index and elevated ozone levels, and highlights an ozone warning in effect for Gyeonggi-do, Chungbuk, and Jeonbuk. The video concludes with the presenter transitioning to the "Economic Time" segment, encouraging viewers to dedicate 30 minutes a day to understanding the economy, and finally, the news anchors wrap up the broadcast, summarizing the key points and preparing for their final remarks. | {
"original": " 8월의 첫날인 오늘 최악의 폭염이 이어졌습니다. 강원도 홍천은 낮기온 41도까지 올랐고요. 서울도 39.6도까지 오르면서 기상관측 이래 가장 높은 기온을 보였습니다. 내일도 전국이 뜨겁겠습니다. 내일 서울 낮기온 39도까지 오르겠고 당분간 더위는 이어지겠습니다. 금요일부터는 점차 기온이 낮아지긴 하겠지만 그래도 낮기온 35도 안팎으로 무덥겠습니다. 아침 기온도 25도를 웃돌면서 밤사이 열대야가 나타나는 곳이 있겠습니다. 내일도 전국이 맑겠습니다. 현재 경기도와 충북, 전북에서는 오존주의보가 발효 중인데요. 내일도 전국이 맑은 가운데 자외선 지수가 강하겠고 오존 농도도 높게 나타나겠습니다. 날씨정보 전해드렸습니다. 하루 30분만 경제타임에 투자하세요. 경제가 한눈에 보입니다. 고맙습니다. 지금까지 뉴스 스토리였습니다.",
"refined": "Today, the worst heat of the first day of August has continued. In Hongcheon, Gangwon-do, the temperature rose to 41 degrees, while in Seoul, it reached 39.6 degrees, the highest temperature since weather records began. Tomorrow, the entire country will be scorching hot, with Seoul's temperature expected to reach 39 degrees and the heat wave persisting for the time being. Although the temperature will gradually drop from Friday, it will still be sweltering, with daytime temperatures around 35 degrees. Morning temperatures will also rise above 25 degrees, with some areas experiencing tropical nights. Tomorrow is expected to be sunny nationwide, with a high UV index and high ozone levels, particularly in Gyeonggi-do, Chungbuk, and Jeonbuk, where an ozone alert is currently in effect. That's the latest weather update.",
"summarized": "South Korea experienced its worst heatwave on the first day of August, with temperatures in Hongcheon, Gangwon-do, reaching 41 degrees and Seoul reaching 39.6 degrees, the highest since weather records began. The heat is expected to continue tomorrow with temperatures in Seoul reaching 39 degrees. Although temperatures will gradually drop from Friday, the country will remain warm with daytime temperatures around 35 degrees and morning temperatures above 25 degrees, with a high risk of tropical nights. The weather is expected to remain clear, with strong sunlight and high ozone levels.",
"translated": "Today, the first day of August, has seen the worst heat wave continue. In Hongcheon, Gangwon-do, the daytime temperature soared to 41 degrees, while Seoul reached 39.6 degrees, marking the highest temperature ever recorded since weather observations began. Tomorrow is expected to be hot nationwide, with Seoul's daytime temperature predicted to rise to 39 degrees, and the heat is likely to persist for the time being. Although temperatures will gradually decrease from Friday, it will still feel sweltering with daytime temperatures hovering around 35 degrees. Morning temperatures will also remain high, exceeding 25 degrees, and tropical nights are expected in some areas. Tomorrow is forecast to be sunny nationwide, with a high UV index and elevated ozone levels, particularly in Gyeonggi-do, Chungbuk, and Jeonbuk, where an ozone warning is currently in effect. That's the latest weather update. Invest just 30 minutes a day in Economic Time, and you'll have a clear view of the economy. Thank you, and that's all for today's News Story.",
"translated_llm": "The worst eruption of August today. The intensity of the flood reached 41°C. Seoul also reached 39.6°C and saw the highest temperature ever observed. Tomorrow the whole country will be hot. Tomorrow Seoul will rise to 39°C and will continue for a few more days.",
"translated_whisper": " The worst of the first day of August has continued today. Hongcheon, Gangwon-do, has risen to a temperature of 41 degrees. Seoul also rose to 38.6, 39.6 degrees, showing the highest temperature since the weather forecast. The whole country will be hot tomorrow. Tomorrow, Seoul's temperature will rise to 39 degrees and the heat will continue for the time being. From Friday, the temperature will gradually drop, but it will still be warm with a temperature of 35 degrees. Morning temperature will rise to 25 degrees and there will be tropical The sun will be strong in the middle of the day and the sun will be high in the morning. I gave you the weather information. Yes, invest in the economy for 30 minutes a day. The economy looks like a single eye. Thank you."
} |
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 ... | [
"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, a... | [
"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... | {
"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"
} | 32 | [
"The news anchors are discussing a recent incident in Bundang, Gyeonggi Province, where a water pipe burst due to the heatwave, leading to a crack in a bridge. The incident occurred during the 10 o'clock interval last night, and the traffic is currently being controlled.",
"The news anchors are continuing their d... | XYrtvA02lKc | A news report is discussing a recent incident in Bundang, Gyeonggi Province, where a water pipe burst due to a heatwave, resulting in a significant crack in a bridge. The incident occurred around 10 o'clock the previous night, and traffic in the area is being controlled. The news anchors are providing updates on the situation, including the management of the cracked bridge and the measures taken to control traffic. The bridge has a visible gap and is cordoned off with yellow barricades and caution tape, indicating restricted access. The road is closed with orange traffic cones, and vehicles are visible in the background, suggesting ongoing traffic control measures. The text on the screen emphasizes that the crack was caused by a water pipe bursting due to the heatwave. Local authorities, including the Bundang police, are involved in managing the situation. The video shows the damaged section of the bridge with noticeable cracks, bulges, and exposed rebar, highlighting the extent of the damage caused by the burst water pipe. The surrounding area includes debris and vegetation, indicating the impact of the incident. Overall, the video depicts the aftermath of the water pipe burst and the efforts to manage the situation and ensure public safety. | {
"original": " 경기도 분당에서 폭염으로 수도배관이 파열되면서 교량에 균열이 발생해 교통이 통제되고 있습니다. 어젯밤 10시 구간의",
"refined": "A burst water pipe caused by a heatwave in Bundang, Gyeonggi Province has led to cracks in the bridge, resulting in traffic control.",
"summarized": "A heatwave in Bundang, Gyeonggi, caused a water pipe to burst, resulting in a crack on a bridge, which has led to traffic control measures being put in place.",
"translated": "In Bundang, Gyeonggi Province, a water pipe burst due to the heatwave, causing a crack in the bridge, and traffic is being controlled. This occurred during the 10 o'clock interval last night.",
"translated_llm": "The city's transportation system was severely damaged by an explosion in the district's main street.",
"translated_whisper": " A tornado has hit the capital of Kyeonggi, where the traffic is under control due to a large number of landslides."
} |
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 ... | [
"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, a... | [
"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... | {
"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"
} | 32 | [
"The image shows a news broadcast from Eunaham News TV with a timestamp of 13:36. The male news anchor in a gray suit is seated in a newsroom filled with digital screens displaying various charts and data. The female news anchor, dressed in a red blazer and black top, is positioned to the right with a background im... | 1038669934375493632 | The video depicts a news broadcast from Eunaham News TV discussing the stable power supply during the summer despite extreme heatwaves. The news anchors present information on the power demand and supply, highlighting that there was no power shortage throughout the summer. The video cuts to a control room where personnel are monitoring and managing the power supply, ensuring its stability. It also shows graphs and data illustrating the power supply and demand trends, with a focus on the reserve rate remaining in double digits for most of the summer, except for a few days. The video emphasizes the successful management of the power supply, with the text "재난수준 폭염에도 올 여름 전력부족 없었다" (No power shortage despite disaster-level heat) consistently displayed. The latter part of the video shows individuals in a control room and a close-up of electricity meters, highlighting the technical infrastructure and precise measurement of electricity usage that contributed to the stable power supply. The original ASR transcript supports the video's narrative, stating that despite the high power demand for air conditioning due to heat waves, the supply reserve rate remained sufficient, with only a few days experiencing a lower reserve rate. Overall, the video and ASR transcript convey the message that the power supply was effectively managed during the extreme heatwaves, ensuring no shortages occurred. | {
"original": " 올여름 재난수준의 폭염으로 냉방 전력 수요가 급증했지만 전력이 부족한 날은 없었던 것으로 나타났습니다. 전력거래소에 따르면 지난 7월과 8월 두 달 동안 공급 예비율은 7월 23일부터 27일, 지난달 13일에서 14일 등 모두 7일간을 제외하고는 모두 두 자릿수 이상을 유지했습니다. 특히 최대 전력 수요가 역대 최고치를 찍었던 7월 24일에는 예비율이 7.7%까지 낮아졌지만 이때도 공급 예비력은 위기경보수 준비단계 기준보다 많은 709만 킬로와트에 달했습니다.",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "This summer, the power demand for air conditioning skyrocketed due to disaster-level heat waves, but there were no days when the power supply was insufficient. According to the power exchange, the supply reserve rate for July and August remained in double digits, except for a total of 7 days, from July 23 to 27 and August 13 to 14. Notably, on July 24, when the maximum power demand reached an all-time high, the reserve rate dropped to 7.7%, but the supply reserve still exceeded the crisis warning preparation stage standard by 709,000 kilowatts.",
"translated_llm": "According to the power exchange, the supply reserve rate for the last two months of July and August remained at or above two digits except for seven days from July 23 to 27, and last month from July 13 to 14, respectively. Especially on July 24, when the maximum power demand recorded its record peak, the reserve rate dropped to 7.7%, but even then the supply reserve power was more than 7.09 million kilowatts above the crisis readiness benchmark.",
"translated_whisper": " The cold power demand has skyrocketed due to the heat wave of the disaster this summer, but there have been no days when power was lacking. According to the power exchange, the supply reserve for the past two months, July and August, has remained two places above the seven days, from July 23 to July 27, and from the 13th to the 14th of last month. In particular, on July 24, when the maximum power consumption was the highest ever, the reserve rate was lowered to 7.7%, but the supply reserve was 709,000 kilowatts more than the standard of the crisis warning preparation stage."
} |
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 ... | [
"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, a... | [
"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... | {
"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"
} | 32 | [
"In the current frame, a scene of an urban street is depicted with several people crossing the street amidst heavy traffic. The image shows a mix of pedestrians and vehicles, including a person riding a motorcycle. The environment suggests a busy, possibly congested area with visible reflections on the wet road sur... | 1026325665824403456 | The video depicts a narrative focused on safety awareness, particularly during extreme weather conditions like heatwaves. It begins by showing scenes of urban streets with heavy traffic, emphasizing the importance of road safety, as indicated by the "안전韩" (Safety Korea) logo. The frames then transition to show individuals crossing streets, walking in parks, and engaging in daily activities, highlighting the need for vigilance in various settings.
As the video progresses, it shifts to a more personal and emotional tone, showing a framed photograph of three individuals, suggesting strong bonds and the importance of checking in on loved ones, especially during hot weather. The setting then changes to an office environment, where a person is shown working at a desk and later sitting in contemplation, possibly reflecting on the importance of safety and the need to be informed about emergency alerts.
The video further emphasizes the theme of safety awareness by displaying a smartphone screen with an emergency disaster alert and a heatwave warning, prompting the viewer to consider the importance of staying informed and checking in on others. The individual in the video is shown engaging in a phone call, likely discussing the emergency alert or ensuring someone's safety, reinforcing the message of concern and care.
Throughout the video, the "안전韩" (Safety Korea) logo and the text "관심이 안전을 지킵니다" ("Concern keeps you safe") are prominently displayed, driving home the importance of vigilance and care in preventing accidents and ensuring safety, especially during extreme weather conditions. The final frame features the logo of the "해양안전부" (Korea Coast Guard), further emphasizing the role of awareness and preparedness in public safety.
Overall, the video conveys a powerful message about the importance of concern, care, and safety awareness, encouraging viewers to be proactive in protecting themselves and others, particularly during emergencies and extreme weather conditions like heatwaves. | {
"original": " 우리가 모르는 사이 누군가는 위험에 빠질 수 있습니다 어? 깜짝이야 긴급재난문자? 그 어느 해보다 뜨거운 올여름 어, 아빠 폭염엔 부모님의 안부를 꼭 물으세요 나의 관심이 누군가의 안전을 지킵니다",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Without our knowledge, someone could be in danger. Oh my, an emergency disaster alert? This summer is hotter than ever. Dad, during a heatwave, be sure to check in on your parents. My concern can help keep someone safe.",
"translated_llm": "Someone we don't know could be in danger. My concern is for someone's safety.",
"translated_whisper": " In the midst of our ignorance, someone could fall into danger. Oh my gosh. Emergency disaster message? This summer is hotter than any other year. Hey, Dad. Please ask your parents for help when you are in a crisis. My interest protects someone's safety."
} |
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 ... | [
"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, a... | [
"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... | {
"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"
} | 32 | [
"The current frame shows raw bacon being placed on a griddle, with pieces spread out. The context seems to be about cooking or food preparation. The visible text in Korean indicates a time and location related to a news broadcast or a cooking show segment.",
"The current frame shows a wedge of cheese with a cream... | 09W_DRDb9HQ | The video appears to be a cooking or food preparation segment, likely part of a news broadcast or television show, where various ingredients such as bacon, cheese, and eggs are being prepared. The scene progresses to show the cooking process, with the ingredients being manipulated and cooked over time. However, the video takes a turn and shifts to an experiment or demonstration involving the melting of chocolate, showcasing how it melts under certain conditions over a period of 15 minutes. The final part of the video seems to introduce a new segment or chapter, titled "Jung Yeong Video," which may be unrelated to the initial cooking or chocolate melting segments, and could potentially focus on a character or topic named Jung Yeong. The original ASR transcript seems to be more focused on a personal or relationship context, discussing someone's new partner and their characteristics, which does not directly relate to the visual content of the video. Overall, the video seems to be a collection of different segments or experiments, with the cooking and chocolate melting demonstrations being the most prominent, followed by an introduction to a potentially new topic or character. | {
"original": " ༼� つ༎ ༼ Yow, Yow, Yow! You got a new girlfriend! And you got a new partner! She's already in a new relationship... So she's got a new relationship... You are all grown up and ready for her... What does her new partner look like? The man from the start. He is a young man... who has a very special relationship with his partner. How can you forget about him even though he is young?",
"refined": "You've got a new girlfriend and a new partner, but she's already in a relationship. So she's got a new partner. You're all grown up and ready for her. What does her new partner look like? He's the man from the start, a young man with a very special relationship with his partner. How can you forget about him despite being young?",
"summarized": "Not Available",
"translated": "Yow, Yow, Yow! You've got a new girlfriend! And you've got a new partner! She's already in a new relationship... So she's got a new relationship... You're all grown up and ready for her... What does her new partner look like? The man from the start. He's a young man... who has a very special relationship with his partner. How can you forget about him even though he's young?",
"translated_llm": "( つ༎ (Yow, Yow, Yow! You got a new girlfriend! And you got a new partner! She's already in a new relationship... So she's got a new relationship... You're all grown up and ready for her... What does her new partner look like? The man from the beginning. He's a young man... who has a very special relationship with his partner. How can you forget about him even though he's young?",
"translated_whisper": " How to make a Thank you for watching."
} |
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 ... | [
"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, a... | [
"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... | {
"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"
} | 32 | [
"In the current frame, a news anchor stands in front of a background featuring an image of a watermelon field. The watermelons on the ground appear to be discarded and rotten, with some cut open, revealing red flesh. The news ticker at the bottom of the screen reads \"수박발 갈아엎어야... 제철 과일 직격탄\" (Translation: \"Waterm... | 8wcKX0LCc08 | The video depicts the impact of a severe heatwave on agricultural crops, particularly watermelons, corn, and cucumbers, in various regions. Reporter Cho Sung-sik presents a news segment from a highland field where watermelons are grown, showing watermelons with rotten flesh, nicknamed "blood watermelons," being discarded due to the heatwave damage. The heatwave has caused water to accumulate inside the watermelons, leading to rot and a significant loss of value, resulting in over half of the crop being discarded. The segment also highlights the struggles of corn crops, which are withering and turning yellow due to the drought, with the fruit not ripening properly, leading some farmers to abandon their harvest. In contrast, cucumbers are overproducing due to the high temperatures, causing prices to plummet and forcing farmers to discard their crop. The video showcases the varying responses of different plants to the extreme weather conditions, emphasizing the complexity of the agricultural challenges posed by the climate. Throughout the segment, reporter Cho Sung-sik explains the situation, providing context to the damaged crops and the economic impact on farmers, while also showing farmers continuing to work amidst the challenges, highlighting their resilience in the face of adverse weather conditions. | {
"original": " 폭염 때문에 힘든 건 사람만이 아닙니다. 과육이 물러진 수박이 산지에서 폐기되는 등 제철과일과 채소 피해도 잇따르고 있습니다. 조성식 기자입니다. 여름철 대표 과일인 수박을 재배하는 고냉지 밭입니다. 농구공만큼 자란 수박들이 바닥에 나뒹굴고 있고 빨간 속살을 드러낸 깨진 수박들은 한켠에 쌓여 그대로 썩어가고 있습니다. 겉은 멀쩡하지만 속은 짙은 붉은색을 띠며 과육이 물러지는 일명 피수박이 된 것입니다. 이처럼 폭염 피해 때문에 상품 가치가 없어지면서 이곳 밭의 경우 생산된 수박의 절반 이상을 폐기 처분했습니다. 뿌리에서 수분이 올라와서 당분은 여기 저장하고 나머지 수분은 증발을 시켜야 하잖아요. 증발시키지 못하니까 수분이 여기 고이는 거죠. 제철을 맞은 다른 농산물들도 상황은 비슷합니다. 폭염과 가뭄이 계속되면서 옥수수 잎은 바짝 말라 누렇게 변했고 열매도 제대로 맺지 못해 아예 수확을 포기했습니다. 반대로 오이는 고온으로 생산량이 급증해 가격이 폭락하자 농민들은 애써 기른 농작물을 모두 버려야 하는 사태가 발생했습니다. MBC 뉴스 조성식입니다.",
"refined": "It's not just people who are suffering from the heatwave. The watermelon that has become rotten due to the heat is being discarded at the farm, and there are also damages to seasonal fruits and vegetables. This is Cho Sung-sik reporting. In this watermelon field, a representative summer fruit, watermelons the size of basketballs are scattered on the ground. The crushed watermelons, with their red flesh exposed, are piled up in one corner, rotting away. Although the surface appears fine, the inside has turned a deep red color, earning them the name \"blood watermelons\". Due to the heatwave, more than half of the produced watermelons have lost their market value and were discarded. Water from the roots is supposed to be stored as sugar, and excess water should evaporate. However, the water can't evaporate, causing it to accumulate. The situation is similar for other crops. As the heat and drought continue, corn leaves have withered and turned yellow, and farmers have given up on harvesting as the fruits are not properly formed. On the other hand, high temperatures have led to an increase in cucumber production, causing prices to plummet, and farmers are forced to dispose of their entire crop. This is MBC News' Cho Sung-sik.",
"summarized": "A severe heatwave is causing significant damage to crops, including watermelons and other seasonal fruits and vegetables. In a watermelon field, over half of the produce was discarded due to the heat, resulting in a loss of product value. The watermelons suffered from a condition known as \"blood watermelon,\" where the inside turns dark red and becomes unsellable. Similarly, corn crops were severely affected, with leaves drying up and turning yellow, and some farmers were forced to abandon harvesting altogether. Conversely, high temperatures led to an overproduction of cucumbers, causing prices to plummet and farmers to discard their entire crop.",
"translated": "The heatwave is not only affecting humans. Watermelons with rotten flesh are being discarded at farms, and other seasonal fruits and vegetables are also suffering damage. I'm reporter Cho Sung-sik. This is a highland field where watermelons, a summer staple, are grown. Watermelons the size of basketballs are scattered on the ground, and broken ones with exposed red flesh are piled up in a corner, rotting away. Although they appear fine on the outside, their insides have turned a deep red, earning them the nickname \"blood watermelons.\" Due to the heatwave damage, the value of these watermelons has been lost, resulting in over half of the produced crop being discarded. Normally, water is absorbed through the roots, stored as sugar, and excess water is evaporated. However, when evaporation is hindered, the water accumulates, causing the flesh to rot. Other crops are facing similar struggles. The prolonged heatwave and drought have caused corn leaves to wither and turn yellow, and the fruit is not ripening properly, leading some farmers to abandon their harvest. In contrast, the high temperatures have led to an overproduction of cucumbers, causing prices to plummet, and farmers are being forced to discard their entire crop. This is MBC News' Cho Sung-sik.",
"translated_llm": "The weather is not just bad for people. The fruits and vegetables are being thrown away from the mountains, and the fruits and vegetables are also being destroyed. The weather is bad for the fields. The summertime representative of the fruits and vegetables is the frozen field where the fruits are grown. The trees are growing as big as a basketball and the broken branches are piled on the ground, exposing the red spots.",
"translated_whisper": " It's not just people who are suffering from the plague. The watermelon that has been crushed by the plague is being discarded in the mountains. There are also seasonal fruits and vegetable damage. Reporter Cho Sung-sik. This is a field of rice fields that grows watermelon, a representative fruit of the summer. Watermelons as big as basketballs are rolling on the ground, and the crushed watermelon, which revealed red flesh, is piled up in one piece and rotting. The surface is fine, but the inside is dark red, and it has become a so-called blood watermelon that is falling. As the value of the product is lost due to the damage caused by the explosion, the half of the produced watermelon was disposed of. The water comes up from the roots, the sugar is stored here, and the water has to evaporate. The water is flowing here because it can't evaporate. The situation is similar for other crops. As the heat and drought continued, the corn leaves dried up and turned yellow, and the fruits were not properly planted, giving up on harvesting. On the other hand, the high temperature increased production and the price fell sharply, and farmers had to throw away all the hard-earned crops. This is MBC News' Cho Sung-sik."
} |
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 ... | [
"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, a... | [
"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... | {
"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"
} | 32 | [
"A man in a suit and sunglasses is holding a gun, extending it forward with a serious expression. The setting appears to be dimly lit with an orange hue, suggesting a tense or dramatic atmosphere.",
"The man in the suit is looking intently to the side, his expression serious. The dim lighting and the orange hue c... | g9CbgCp52qc | The video depicts a complex narrative that intertwines scenes of action, contemplation, and city management efforts amidst a heatwave. It begins with a man in a suit, often appearing serious and contemplative, engaged in various settings that range from dimly lit, dramatic atmospheres to more formal and serene environments like a candle-lit church. His actions and expressions suggest a deep involvement in significant conversations, decisions, or possibly investigations, highlighted by his gestures, intense gazes, and interactions with other characters.
The video also shifts to scenes that imply a sense of urgency and conflict, including martial arts fights in dimly lit warehouses, altercations on rooftops, and intense, determined expressions of characters in motion. These scenes are juxtaposed with moments of reflection and planning, indicating a narrative that balances action with strategic thinking.
Notably, the video incorporates elements that suggest a broader context, such as the mention of Seoul City's efforts to combat the heatwave by spraying water on roads, which is directly referenced in the ASR transcript. This aspect of the video highlights the city's proactive approach to managing high temperatures and ensuring public safety, providing a contrast to the more personal, intense narratives of the characters.
The inclusion of a promotional image for a cinematic event and scenes of dramatic explosions and action sequences suggests that the video may be a trailer or preview for a film that combines elements of action, drama, and possibly social commentary on urban challenges like heatwaves.
Overall, the video presents a multifaceted narrative that blends personal dramas, action sequences, and social issues, set against the backdrop of a city navigating the challenges of a heatwave. | {
"original": " 연일 불볕더위가 기승을 부리는 가운데 서울시가 도로에 물을 뿌려서 열을 식히는 작업을 강화합니다. 서울시는 낮 최고기온이 32도 이상으로 오르면 폭염에 노출될 가능성이 큰 정오부터 오후 4시까지 중앙버스 전용차로를 중심으로 살수차량 운행을 확대한다고 밝혔습니다. 서울시는 여름철 도로 물뿌리기로 미세먼지를 제거하는 동시에 온열 질환 등 폭염으로 발생할 수 있는 안전사고도 방지할 수 있길 기대하고 있습니다. 촬영기자1호, 박진주 기상캐스터",
"refined": "Seoul City is taking measures to cool down the roads as the heatwave intensifies. If the daytime high temperature exceeds 32 degrees Celsius, the city plans to expand the operation of water-spraying vehicles, primarily along the central bus-only lanes, from noon to 4 pm. Seoul expects this initiative to not only remove fine dust but also prevent heat-related illnesses and safety accidents.",
"summarized": "Seoul City is taking measures to combat the heatwave by spraying water on roads to cool them down. When the temperature exceeds 32 degrees Celsius, the city will expand the operation of water sprinkler vehicles, focusing on central bus-only roads, from noon to 4 pm. This effort aims to remove fine dust and prevent heat-related safety accidents and illnesses.",
"translated": "In the midst of the scorching heat, Seoul City is strengthening its efforts to cool down the roads by spraying water. The city announced that it will expand the operation of water-spraying vehicles, focusing on central bus-only lanes, from noon to 4 pm when the maximum temperature is expected to rise above 32 degrees Celsius. By doing so, Seoul City hopes to not only remove fine dust but also prevent heat-related illnesses and safety accidents that can occur due to the heatwave.",
"translated_llm": "In the midst of the annual sunrise, Seoul city is spraying water on the roads to increase the heat dissipation. Seoul city is expanding the number of vehicles running on the Central Buses from noon to 4 pm, with the possibility of exposure to the eruption if the maximum temperature rises to 32 degrees or above. Seoul city is also hoping to eliminate micro-dust while eliminating the heat illness and preventing safety accidents that can occur from the eruption.",
"translated_whisper": " Seoul City is forcing the road to cool down as the heat is rising. Seoul City said that the heat will rise to more than 32 degrees Celsius during the day and the possibility of exposure to a fire will be high from noon to 4 p.m. It said that the operation of the hydrogen vehicle will be expanded around the central bus-only road. Seoul is hoping to prevent safety accidents that can occur due to heat and heat diseases while removing fine dust with summer road water sprinklers."
} |
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 bundli... | [
"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... | [
"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 bra... | {
"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"
} | 32 | [
"The image depicts a residential area with multiple high-rise buildings lined up on both sides of a pathway. The scene appears to be covered in a light layer of fog, giving it a somewhat hazy appearance. In the foreground, there is a park area with various trees and pathways, indicating a planned residential commun... | 1346958705632296961 | The video depicts a serene residential area with tall buildings and a park, set in a colder season with bare trees and pathways. A gentle breeze is present throughout, causing slight movement in the trees. The scene remains calm and tranquil, with a light fog partially obscuring the trees and pathways in the initial frames. A person is seen walking along a path in some frames, adding activity to the otherwise peaceful setting. A red and white vehicle is parked near the pathways, and building numbers such as "107" and "129" are visible, providing context to the residential setting. The overall atmosphere is peaceful and serene, consistent with the description of a sunny morning with a gentle breeze. | {
"original": " Today is a sunny morning. The green wind blows quickly so that I will get the wind through the trees.",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Today is a sunny morning. The gentle breeze blows quickly through the trees.",
"translated_llm": "Today is a sunny morning. The green wind blows quickly so that I will get the wind through the trees.",
"translated_whisper": " Good Morning! Today let's have a little break and talk about this video."
} |
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 bundli... | [
"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... | [
"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 bra... | {
"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"
} | 32 | [
"The current frame shows a landscape with the sun partially obscured by smoke or fog, indicating a hazy atmosphere likely due to cold weather. The scene includes dry grass in the foreground and hills or mountains in the background. The text at the bottom of the image suggests that a cold wave has begun since the en... | 2qCY0mBeZy4 | The video depicts the severe and prolonged cold wave that affected South Korea, particularly in areas such as Gangwon-do, Cheorwon, and Seoul. The temperature dropped significantly, with lows of -20 degrees Celsius in Gangwon-do's mountainous areas, -15.9 degrees Celsius in Cheorwon, and -9.8 degrees Celsius in Seoul. The cold wave is expected to continue until next week, with a brief rise in temperature over the weekend, followed by a sharp decline. A strong cold wave is forecasted to hit in the second half of next week, with temperatures in Seoul potentially dropping to -15 degrees Celsius, making it the coldest winter so far.
The video shows the impact of the cold wave on daily life, with people bundled up in heavy winter clothing, and workers taking measures to address the challenges posed by the freezing temperatures, such as breaking ice in pipes and containers. The cold wave also affected transportation, with snow-covered roads and icy conditions making it difficult for vehicles to navigate.
The Korea Meteorological Administration warned of potential damage to facilities such as frozen pipes and vinyl greenhouses due to the cold wave and heavy snowfall, and urged people to take precautions in advance. The video also shows children playing in the snow, highlighting the resilience and adaptability of the public in the face of harsh weather conditions.
Overall, the video provides a comprehensive overview of the severe and prolonged cold wave that affected South Korea, its impact on daily life, and the measures taken to mitigate its effects. The video serves as a warning, emphasizing the need for preparedness and precautions in the face of extreme weather conditions. | {
"original": " 지난 연말부터 시작된 한파의 기세는 여전했습니다. 강원도 산간은 영하 20도, 철원 영하 15.9도, 서울도 영하 9.8도를 기록했습니다. 이번 한파는 풀리지 않고 다음주까지 이어질 것으로 보입니다. 주말부터 다음주 초반까지 기온이 조금 오르지만 이후 다시 기온이 급강화하기 때문입니다. 특히 다음주 후반에는 서울기온이 영하 15도까지 떨어지는 강력한 한파가 맹위를 떨칠 전망입니다. 올겨울 들어 가장 강력한 추위입니다. 북극 한파가 새해 첫날부터 열흘 이상 이어지는 것입니다. 서해안의 눈도 잠시 약해졌다가 다음 주 중반 이후에는 눈폭탄으로 쏟아질 것으로 우려됩니다. 기상청은 이번 한파 폭설로 동파와 축사 비닐하우스 붕괴 등 시설물 피해가 속출할 것이라며 사전 대비를 당부했습니다",
"refined": "The cold wave that began at the end of last year continues unabated. Temperatures in the mountains of Gangwon-do dropped to minus 20 degrees Celsius, minus 15.9 degrees Celsius in Cheorwon, and minus 9.8 degrees Celsius in Seoul. This cold wave is expected to persist until next week, with temperatures rising slightly over the weekend and early next week before plummeting again. A strong cold wave is forecasted for the latter half of next week, with temperatures in Seoul potentially plummeting to minus 15 degrees Celsius. This will be the coldest period this winter, with the Arctic cold wave persisting for over 10 days since the start of the new year. Meanwhile, snow in the western coastal area is expected to temporarily weaken before intensifying into a heavy snowfall in the middle of next week. The weather agency has urged precautions against potential damage to facilities, including pipes bursting and vinyl greenhouses collapsing, due to the cold wave and heavy snow.",
"summarized": "A severe cold wave that began at the end of last year is expected to continue until next week, with temperatures in some areas of South Korea reaching as low as -20°C. Despite a brief rise in temperature over the weekend, the cold snap is expected to intensify again, with Seoul potentially dropping to -15°C by the end of next week. This will be the strongest cold wave of the winter, with the North Pole cold wave lasting over 10 days from the start of the new year. Additionally, heavy snow is forecast for the west coast, which may cause damage to facilities such as greenhouses and vinyl houses. The weather office is urging people to take precautions in advance to mitigate potential damage.",
"translated": "The cold wave that started at the end of last year continued to persist. The temperature in the mountainous areas of Gangwon-do dropped to below -20 degrees Celsius, while Cheorwon recorded a low of -15.9 degrees Celsius, and Seoul reached a low of -9.8 degrees Celsius. This cold wave is expected to continue until next week without letting up. Although the temperature will rise slightly from the weekend to the beginning of next week, it will plummet again afterwards. In particular, a strong cold wave is forecasted to hit in the second half of next week, with temperatures in Seoul expected to drop to as low as -15 degrees Celsius, making it the coldest winter so far. The North Pole cold wave is expected to last for over 10 days starting from New Year's Day. The snowfall on the west coast is expected to subside temporarily but may intensify again in the middle of next week, with concerns of a heavy snowfall. The Korea Meteorological Administration has warned of potential damage to facilities such as frozen pipes and vinyl greenhouses due to the cold wave and heavy snowfall, and urged people to take precautions in advance.",
"translated_llm": "The winds of the Han wave, which started at the end of last year, are still strong. The intensity of the Han wave in the mountains is below 20 degrees, the iron core is below 15.9 degrees, and the Seoul side is below 9.8 degrees. The winds of this one are likely to continue without breaking until next week. The temperatures will rise slightly from the weekend to the beginning of the next week, but then intensify again.",
"translated_whisper": " The trend of the Korean wave, which started last year, remained. The mountains in Gangwon-do are below 20 degrees Celsius, Cheorwon is below 15.9 degrees Celsius, and Seoul is below 9.8 degrees Celsius. This Korean wave will not be released and will continue until next week. The temperature rises slightly from the weekend to the beginning of next week, but the temperature rises rapidly again. Especially in the second half of next week, the strong cold wave, which falls to minus 15 degrees in Seoul, is expected to break through the cold. This is the strongest cold this winter. In the event that the cold wave does not become more intense, the second wave will come down with an additional one giga. Therefore, next week, the cold will be much stronger than it is now. The North Pole cold wave will continue for more than 10 days from the first day of the new year. The snow in the west will be weakened for a while, and then it will be poured with a snow bomb in the middle of next week. The weather office has promised to prepare in advance for the damage to the facilities such as the collapse of the Dongpa and the Chuksa vinyl house with the storm and the explosion."
} |
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 bundli... | [
"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... | [
"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 bra... | {
"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"
} | 32 | [
"The current frame shows a news reporter standing near a river, likely delivering a weather update. The setting appears cold, with ice on the river and snow visible along the riverbank, indicating a severe cold wave. The reporter is discussing the drastic temperature drop, with temperatures plummeting to -11 degree... | QmU3W9ffZeM | A news reporter stands near a river, delivering a weather update on an impending severe cold wave. The temperature in Seoul is expected to drop drastically, plummeting to -11 degrees tomorrow morning, -15 degrees on Thursday, and reaching a low of -17 degrees on Friday, marking the peak of the cold wave. A cold wave warning will be issued for specific regions, including parts of Gyeonggi-do, Gangwon-do, Chungcheong-do, and Gyeongsangbuk-do, while a cold wave alert will be expanded to other areas. Snow clouds will move in, bringing heavy snowfall of over 30 cm to the West Coast and over 50 cm to mountainous areas in Jeju Island, with snowy conditions persisting until Friday in the Chungnam region and until Sunday in the Honam and Jeju Island areas. The reporter advises viewers to take precautions in advance, emphasizing the severity of the cold wave and the need for preparation. The weather update is accompanied by visuals of a cold and snowy environment, highlighting the seriousness of the forecast. | {
"original": " 기온이 마치 롤러코스터처럼 떨어질 모양입니다. 얼마나 추워질지 벌써부터 겁도 나는데요. 일단 내일 아침 서울의 기온은 영하 11도까지 떨어지겠고요. 목요일부터는 찬 바람이 더욱 강해지면서 아침 기온이 영하 15도로 전날 낮 대비 13도가량 기온이 급강하겠습니다. 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 9시 일기에서는 경기 동부와 북부, 강원 내륙과 충청도, 경북 곳곳에 한파경보가 발효되겠고요. 서울 등 그 밖의 수도권과 충청도, 전북 내륙 지역까지 한파주의보가 확대발효되겠습니다. 눈구름도 함께 몰려오겠습니다. 내일 퇴근길 무렵에 서울 등 서쪽 지역부터 눈이 내리기 시작해 밤사이에 전국으로 확대될 텐데요. 특히 서해안 지역으로는 최고 30cm 이상, 제주 산지에는 최고 50cm 이상의 폭설이 우려됩니다. 충남 지역에는 금요일까지, 호남과 제주도는 일요일까지 눈이 길게 이어지겠습니다. 날씨였습니다.",
"refined": "The weather will be like a roller coaster. I'm already worried about how cold it will get. Tomorrow morning, Seoul's temperature will drop to -11 degrees. From Thursday, the cold wind will strengthen, and the morning temperature will drop to -15 degrees, a 13-degree drop from the previous day. The daytime temperature will remain at -11 degrees. Next Friday morning, the temperature will hit -17 degrees, the peak of this cold wave. It's best to prepare in advance. Accordingly, a cold wave warning will be issued for the eastern and northern parts of Gyeonggi-do, the inland areas of Gangwon-do and Chungcheong-do, and parts of Gyeongsangbuk-do at 9 pm today. A cold wave alert will also be expanded to the rest of the metropolitan area, Chungcheong-do, and the inland areas of Jeollabuk-do. Snow clouds will move in together. Snow will start falling in the Seoul area and western regions during the evening commute tomorrow and will spread nationwide by night. Heavy snowfall of over 30 cm is expected in the West Coast region, and over 50 cm in the mountainous areas of Jeju Island. The snow is expected to continue until Friday in the Chungnam region and until Sunday in the Honam region and Jeju Island.",
"summarized": "A severe cold wave is forecasted to hit the region, with temperatures expected to drop significantly. Seoul's temperature will drop to -11°C tomorrow morning and -15°C by Thursday, with a further drop to -17°C on Friday. The cold weather will be accompanied by strong winds and snow, with the West Coast region expecting over 30cm of snow and Jeju Island over 50cm. Snow is expected to start falling in western regions, including Seoul, during tomorrow's evening commute and will spread nationwide by night. The cold weather is expected to persist, with the Chungnam region experiencing snow until Friday and Honam and Jeju Island until Sunday.",
"translated": "The temperature is expected to drop drastically, like a roller coaster. I'm already getting worried about how cold it will get. Tomorrow morning, the temperature in Seoul will plummet to -11 degrees. From Thursday, the cold wind will intensify, causing the morning temperature to drop to -15 degrees, a 13-degree decrease from the previous day. The daytime temperature will remain at -11 degrees. The following morning, Friday, the temperature will hit a low of -17 degrees, marking the peak of this cold wave. It's advisable to take precautions in advance. Accordingly, a cold wave warning will be issued for the eastern and northern parts of Gyeonggi-do, the inland areas of Gangwon-do and Chungcheong-do, and parts of Gyeongsangbuk-do at 9 pm tonight. A cold wave alert will also be expanded to include the rest of the metropolitan area, Chungcheong-do, and the inland regions of Jeollabuk-do. Snow clouds will also move in, with snow expected to start falling in the Seoul area and western regions during tomorrow's commute, spreading nationwide by night. Heavy snowfall of over 30 cm is predicted for the West Coast, and over 50 cm for mountainous areas in Jeju Island. The snowy weather will persist until Friday in the Chungnam region, and until Sunday in the Honam and Jeju Island areas. That's the latest weather update.",
"translated_llm": "The temperature will drop like a roller coaster. I'm already scared of how cold it will get. Once tomorrow morning the temperature in Seoul will drop to 11 degrees. From Thursday, the cold wind will be even stronger as the morning temperature will rise to 13 degrees. One day the temperature will stay at 11 degrees. The next day Friday morning the temperature will be at 11 degrees.",
"translated_whisper": " The weather is going to be like a roller coaster. I'm already scared of how cold it will get. First of all, Seoul's temperature will drop to 11 degrees tomorrow morning. From Thursday, the cold wind will become stronger, and the morning temperature will be 15 degrees below zero. The temperature will be 13 degrees stronger than the day before. The hot and cold temperature will stay at 11 degrees below zero. The next Friday morning temperature will be 17 degrees below zero. It will be the key to this wave. It would be better to do it in advance than one wave. Accordingly, the weather forecast will be expanded to the north and south of Gyeonggi-do, the north and south of Gyeongbuk, and the south and south of Gyeongbuk. Seoul and other cities will be expanded to the north and south of Gyeongbuk. Snow clouds will also come together. Tomorrow, snow will start to fall from the west, including Seoul, and it will expand throughout the country in the middle of the night. In particular, the West Coast region will have a snowstorm of more than 30 centimeters and Jeju Island will have a snowstorm of more than 50 centimeters. The weather forecast for the day will continue to be long until Friday in the Chungnam area and Sunday in Honam and Jeju Island. It was the weather."
} |
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 bundli... | [
"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... | [
"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 bra... | {
"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"
} | 32 | [
"A weather presenter stands in front of a snowy mountain backdrop, reporting on the day's weather. The graphic on the screen shows temperature forecasts, indicating a drop from -9.6°C to -4.0°C, with a peak of -6.2°C. The presenter mentions a heavy snow warning for Jeju Island, Gochang, Seocheon, and parts of Jangs... | G7XGz4uL6mI | A weather presenter delivers a forecast update, warning of a heavy snow warning in effect for Jeju Island, Gochang, Seocheon, and parts of Jangseong due to a low-pressure system passing through the north. The presenter advises viewers to drive safely as snow is expected to start falling from the west coast and reach the central region by the evening commute, causing slippery road conditions. The forecast indicates a significant temperature drop, with temperatures ranging from -9.6°C to -4.0°C, and a peak of -6.2°C. Throughout the video, the presenter emphasizes the importance of safe driving due to the expected slippery conditions, providing updates on the snowfall regions and times, and highlighting the affected areas on a map. The video focuses on informing viewers about the weather conditions and providing necessary precautions to ensure safe travel. | {
"original": " 기상캐스터 배혜지 제주도와 고창, 서천과 일부 장성 지역에는 대설주의보가 발효 중입니다. 오늘 곳곳에 눈 소식이 들어 있는데요. 북쪽을 지나는 기압골의 영향으로 서해안부터 점차 눈이 내리기 시작해서 중부지방에는 퇴근길 무렵 눈이 내리겠습니다. 도로가 미끄러운 곳이 많으니까 안전운전하셔야겠습니다. 날씨 전해드렸습니다.",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "A heavy snow warning is in effect for Jeju Island, Gochang, Seocheon, and parts of Jangseong. There are reports of snow in various areas today. Due to the influence of a low-pressure system passing through the north, snow will start falling from the west coast and will reach the central region by the evening commute. Please drive safely as there are many slippery roads. That's the latest weather update.",
"translated_llm": "In the northern part of the country, due to the influence of the air pressure, it will start to snow gradually from the west coast and in the central region it will snow on the way out. The roads are slippery so you should be safe. The weather has been reported.",
"translated_whisper": " The temperature will rise from the afternoon. The temperature will rise to 1 degree in Seoul and 5 degrees in the middle of the day. It will be cold until morning. The weather will be warm and fine dust will be present. The temperature will rise in the western region and the fine dust will be present in the area. The weather forecast is in place in Jeju Island, Gochang, Seochon and some areas of Jangseong. There are snowy weather reports in many places today. The snow is starting to fall from the west coast due to the influence of the snowfall passing through the north. It will snow in the central region as soon as you get off work. There are many slippery roads, so you should drive safely. I told you the weather."
} |
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 bundli... | [
"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... | [
"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 bra... | {
"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"
} | 32 | [
"The current frame shows a snowy urban street scene at night. Snow is actively falling, covering the ground and buildings. Bright lights from signs, including one that reads \"JUNO HAIR,\" illuminate the area, creating a contrast against the dark, snowy backdrop. The setting suggests a quiet, possibly residential a... | 1346762056142188546 | The video depicts a serene and tranquil urban street scene at night, set against the backdrop of a heavy snowstorm. The street, lined with illuminated signs including one that reads "JUNO HAIR," remains mostly empty throughout, emphasizing the quiet and peaceful atmosphere. Snow falls continuously, blanketing the ground and buildings in a thick layer, creating a picturesque winter setting. The only signs of life are two people seen walking in the distance in the later frames, subtly adding a sense of activity to the otherwise still and peaceful scene. The audio transcript, though incomplete, suggests an introductory or narrative element that is not fully captured in the visual descriptions provided. Overall, the video captures the intense yet serene beauty of a snow-covered urban landscape on a quiet night. | {
"original": " Hello! I will introduce you to the",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Hello! I'll introduce you to the",
"translated_llm": "I'll introduce you to the",
"translated_whisper": " If you are new to my channel, welcome!"
} |
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 bundli... | [
"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... | [
"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 bra... | {
"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"
} | 32 | [
"A person is standing on a snowy sidewalk by a pedestrian crossing, holding a sign or a board, possibly directing or guiding. The background features a street with cars and construction cones, suggesting some roadwork or traffic control. Trees without leaves indicate it is winter. The setting appears to be an urban... | 1346961474866925570 | The video appears to depict a scene at the South Coast Fishing Port, specifically highlighting an area with ongoing roadwork or construction, as indicated by the presence of orange traffic cones and signs. The setting is urban, with tall buildings and leafless trees, suggesting it is winter.
Initially, a person is seen standing on a snowy sidewalk, holding a sign or board, possibly directing or guiding. As the video progresses, more individuals appear, also holding signs with Korean text, suggesting a demonstration or public awareness event related to the fishing port. These individuals are seen standing at a pedestrian crossing, walking, and addressing passersby or the public.
The video also focuses on the entrance of a building marked with a blue banner reading "South Coast Fishing Port, South Coast Fishing Port, Sebaan Pallar," where security or staff members are present, holding signs. The area is restricted due to construction or traffic control, with orange cones and signs in place.
Throughout the video, the urban setting remains consistent, with an emphasis on the activities around the South Coast Fishing Port. Vehicles, including a silver van, are seen parked on the side of the street, with some indicating ongoing work or delivery activities. The overall atmosphere suggests a quiet, winter scene with minimal pedestrian or vehicular activity, apart from the demonstration or awareness event and the construction work. | {
"original": " 南海岸沙漁港 南海岸沙漁港 Sebaan Pallar",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "South Coast Fishing Port, South Coast Fishing Port, Sebaan Pallar",
"translated_llm": "South Coast fishing port South Coast fishing port Sebaan Pallar",
"translated_whisper": " It's been a while since I've been on a bus."
} |
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 bundli... | [
"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... | [
"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 bra... | {
"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"
} | 32 | [
"The current frame depicts a news studio with two news anchors standing on a stage, presenting a segment. The studio backdrop shows a nighttime cityscape, and there is a large screen displaying text in Korean. The news anchors appear to be discussing recent events, likely related to the ongoing cold weather trend m... | PxIbNGPQYw8 | The video depicts a news segment covering the extreme cold weather trend and its impact on daily life. Two news anchors present the segment, discussing how people are sharing proof of the cold on social media platforms, showcasing everyday items that have frozen due to the temperatures dropping below -10 degrees. The anchors highlight the "frozen" series trend, which began in 2018, where people share photos and videos of items like iced coffee, water, and cosmetics that have frozen, often with humorous reactions. The segment also reports on the significant increase in damage from frozen pipes and other cold-related issues, with over 5,000 cases reported. Throughout the video, various visuals illustrate the trend, including frozen drinks, bottled water, and other items, as well as people's reactions to the cold. The news anchors note that despite the challenges posed by the cold weather, people are using humor to cope, turning the experience into a game and sharing their unique cold-weather experiences on social media. The overall tone of the segment is informative, showcasing the widespread impact of the cold wave on the country, while also highlighting the resilience and creativity of people in the face of extreme weather conditions. | {
"original": " 지금까지 생생 지구촌이었습니다. 얼어붙는 강추위가 지속되면서 SNS 같은 온라인상에서 강추위를 인증하는 열풍이 불고 있는데요. 얼어서 시리즈 영상으로 함께 보시겠습니다. 얼음이 들어간 아이스커피가 진짜 꽁꽁 얼어붙었습니다. 차 안에 둔 물티슈도 또 생수도 베란다에 두었던 콜라까지 영하 10도를 밑두는 한파가 모든 걸 얼어붙게 만든 얼어서 시리즈의 인증 장면들인데요. 이런 것도 얼어붙었다라며 저마다 다른 추위 인증샷을 올리는 겁니다. 이 얼어서 시리즈의 유행, 연일 기록적인 한파가 이어지던 지난 2018년에 시작됐는데요. 당시 누리꾼들은 정수기 물과 잼, 커피, 또 화장품 등 뜻밖의 물건이 얼어버린 모습을 공유하면서 당황스러움을 감추지 못했습니다. 중앙재난안전대책본부에 따르면 어제 오후 4시 30분 기준 수도 계량기 등 동파 피해만 5000건이 넘게 발생할 만큼 추위는 전국을 꽁꽁 얼어붙게 만들고 있는데요. 이는 전날보다 2배 이상으로 증가한 수치라고 합니다. 강추위를 일종의 놀이로 이겨내는 재치가 무척 재미있는 것 같습니다. 놀이화를 하면 극복할 수 있는 것들이 참 많아지는 것 같습니다. 맞습니다.",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "We've had a lively coverage of the global village so far. As the freezing cold persists, a trend has emerged on social media platforms like SNS, where people are sharing proof of the cold. Let's take a look at a series of videos showcasing the freezing temperatures. Even iced coffee with ice inside has completely frozen. Water and wet tissues left in the car, as well as bottled water and cola placed on the balcony, have all frozen due to the cold wave that has dropped temperatures below -10 degrees. People are sharing their own unique cold-weather experiences, posting photos and saying, \"This has frozen too.\" This \"frozen\" series trend began in 2018, when record-low temperatures were being reported daily. At the time, people shared surprising photos of everyday items like water, jam, coffee, and cosmetics that had frozen, and their reactions were priceless. According to the Central Disaster and Safety Countermeasures Headquarters, as of 4:30 pm yesterday, there were over 5,000 reported cases of damage from frozen pipes and other cold-related issues, more than double the number from the previous day. The cold is indeed making the entire country feel frozen. It's interesting to see how people are using humor to cope with the cold, turning it into a game. By doing so, they're able to overcome the challenges posed by the cold weather.",
"translated_llm": "So far, it has been a living earth village. While the freezing precipitation continues, a heat wave is blowing to certify precipitation online such as SNS. Let's watch a series of ice videos together. The ice inside the ice sheet really froze. The water in the car and the fresh water on the balcony, even the cola that was placed on the balcony, are under 10 degrees. This is the scene of the certification of a series of ice waves made all freeze.",
"translated_whisper": " Hello everyone. Today, we will be with the new star of the show, the new star of the show, Na Kyung Chul. Hello. Let's start with the first keyword. The Korean Wave has been promoting the series called, The Frozen. The Korean Wave has been promoting the series for a long time. What is the series called, The Frozen? The Korean Wave has been promoting the series called, The Frozen on social media. There are so many interesting things going on these days. As the cold weather continues to freeze the river and the sea, there is a craze for proofing the cold weather on online like SNS. Let's watch the video of the series. The iced coffee that was filled with ice really froze. Even the wet tissues in the car, the water, the coke on the veranda, and the cold series of proofing that the cold wave that was below 10 degrees made everything freeze. They are posting different cold proof shots of themselves saying that this is frozen, too. The trend of the Frozen series began in 2018, when a record-breaking wave of the year was going on. At that time, the tourists shared the unexpected things such as water, jam, coffee, and cosmetics, and they couldn't hide their embarrassment. According to the Central Disaster Safety Initiative, there were over 5000 cases of East-East damage, such as water-borne damage, at 4.30 p.m. yesterday afternoon. The cold is making the whole country frozen. This is a number that has increased by more than twice as much as the previous day. It seems that it is very interesting to overcome the strong cold with a kind of game. I think there are a lot of things that can be overcome by playing. That's right."
} |
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 citi... | [
"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 flo... | [
"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 duri... | {
"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"
} | 32 | [
"In the current frame, a news anchor is standing in front of a backdrop showing a night-time cityscape with a brightly lit bridge. The backdrop suggests an urban setting, and the illuminated bridge adds a sense of depth to the background. The text on the screen includes the name of the news channel, \"연합뉴스TV,\" and... | SPZkwqJerMU | The video reports on the aftermath of the 2020 flood disaster in Korea, where residents suffered significant damage due to poor dam and waterway management. The Central Environmental Dispute Resolution Committee concluded the dispute resolution process between flood victims, the government, local authorities, and the Korea Water Corporation, resulting in approximately 1,483 million won in compensation for the affected residents. The video footage shows extensive debris and damage, including destroyed buildings, rubble, and scattered household items, highlighting the severity of the flood's impact. News anchors from 연합뉴스 TV (Yonhap News TV) present the news, summarizing the resolution and compensation efforts. The video also displays aerial views of flooded areas, with trucks navigating through muddy waters, and scenes of ongoing cleanup efforts, emphasizing the extensive damage and the challenges faced by the affected communities. Overall, the video documents the significant impact of the 2020 flood and the measures taken to compensate and support the affected residents. | {
"original": " 2020년 여름 기나긴 장마로 큰 홍수 피해를 당한 주민들이 총 1,480여 건의 보상금을 지급받게 됐습니다. 환경부 소속 중앙환경분쟁조정위원회는 홍수 피해를 본 주민들과 정부, 지자체 및 한국수전공사 간의 분쟁조정 절차가 최근 마무리됐다고 밝혔습니다. 조정위는 관계부처 원인 조사에서 댐 하천을 미흡하게 관리한 사실이 드러난 점 등을 고려했다고 밝혔습니다.",
"refined": "In the summer of 2020, residents who suffered heavy flooding received a total of 1,480 compensation claims. The Central Environmental Dispute Resolution Committee under the Ministry of Environment announced that the dispute resolution process between the flood-affected residents, the government, local authorities, and the Korea Water Corporation has recently been completed. The committee considered the fact that the dam and river were poorly managed, as revealed in the investigation by relevant authorities.",
"summarized": "Residents affected by severe flooding in the summer of 2020 have received a total of approximately 1,480 compensation payments. The Central Environmental Dispute Mediation Committee facilitated the dispute resolution process between the affected residents, the government, local authorities, and the Korea Water Corporation, considering factors such as inadequate dam management by the relevant parties.",
"translated": "In the summer of 2020, residents who suffered significant flood damage due to the prolonged rainy season received a total of approximately 1,480 compensation payments. The Central Environmental Dispute Resolution Committee, affiliated with the Ministry of Environment, announced that the dispute resolution process between flood victims, the government, local authorities, and the Korea Water Corporation has recently been concluded. The committee considered the findings of the relevant authorities' investigation, which revealed that the poor management of the dam and waterways was a contributing factor.",
"translated_llm": "The Environment Ministry's Central Environmental Dispute Resolution Committee said that the dispute resolution procedure between the flood victims and the government, the local authorities and the Korean Water Corporation has recently been completed. The coordinating body said that it has considered the fact that the investigation of the relevant authorities has revealed that the water source was poorly managed.",
"translated_whisper": " In the summer of 2020, residents who suffered heavy rain and water suffered a total of 1480 won in compensation. The Central Environmental Disaster Control Commission, which is a member of the Ministry of Environment, said that the disaster control process between the residents who suffered heavy rain and water and the government, the local government and the Korea Water and Water Corporation has recently been completed. The administration said it considered the fact that the dam was poorly managed by the Ministry of Revenue."
} |
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 citi... | [
"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 flo... | [
"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 duri... | {
"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"
} | 32 | [
"The image shows an aerial view of a flooded residential or commercial area. Several buildings and vehicles are partially submerged in muddy water. The scene suggests severe flooding, likely from a heavy rainfall or natural disaster. The text \"뉴스1TV\" indicates that this footage is from a news broadcast by a TV st... | mtt7-wXFwxI | A severe flood event is depicted in the video, with footage showing widespread flooding in various areas, including Gwangju and other locations in South Korea. The floodwaters are shown to be high, with many buildings and vehicles partially or fully submerged. Rescue operations are underway, with individuals wearing life jackets and helmets navigating inflatable boats to transport people through the flooded areas. The video also shows people walking through flooded streets, holding umbrellas, and attempting to navigate the challenging conditions. Emergency services, including the Jeonnam Fire Department, are involved in the response efforts. The flood is reported to have caused significant damage and disruption, with over 300 residents displaced. The video is part of a news broadcast by News1TV, providing coverage of the flood event and its impact on the affected communities. | {
"original": " Teksting av Nicolai Winther",
"refined": "Welcome to the channel.",
"summarized": "Not Available",
"translated": "Text by Nicolai Winther",
"translated_llm": "Text by Nicolai Winther",
"translated_whisper": " Welcome to the channel! Today we will make a special special special special special special special special special special special special special"
} |
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 citi... | [
"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 flo... | [
"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 duri... | {
"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"
} | 32 | [
"A news anchor is presenting a report in a studio, with a backdrop showing a foggy or smoky scene. The text on the screen indicates that 22 people have died and 4 are missing due to floods caused by heavy rainfall in North Korea. The news ticker at the bottom of the screen mentions the flood situation and provides ... | 5j_NI2eL3HA | A news report describes the aftermath of severe flooding in North Korea, caused by heavy rainfall, which has resulted in 22 deaths and 4 people missing. The flooding has destroyed or flooded over 8,000 homes and significantly damaged farmland. The report shows a news anchor presenting the information in a studio, with a backdrop of foggy or smoky scenes, and later cuts to footage of individuals in rain gear and masks assessing the damage and assisting with relief efforts. The video also depicts bulldozers navigating through flooded areas, pushing debris and clearing roads, highlighting the ongoing recovery efforts. The scenes show turbulent, muddy waters, emphasizing the challenging conditions faced by rescue workers and affected communities. Throughout the video, the text on the screen consistently reports the number of casualties and the extent of the damage, underscoring the severity of the disaster. The footage of flooded rivers, roads, and paths further emphasizes the impact of the flooding on residential areas and farmland. Overall, the video portrays the devastating effects of the floods in North Korea and the ongoing efforts to manage the aftermath and provide relief to those affected. | {
"original": " 남쪽 러시아에서는 화체화 확산 시스템을 지키고 있어 전해드렸습니다. 전해 주신 여러분 고맙습니다. 북한에서 집중 호우로 발생한 홍수에 22명이 숨지고 4명이 실종됐습니다. 국제적십자사연맹은 미국의 소리 방송에 보낸 이메일에서 북한조선적십자회의와 국가비상재해위원회에 따르면 시민가가 깎은 홍수로 인해 22명이 사망하고 4명이 실종됐다고밝혔습니다. 연안은 북한에서 홍수로 주택 8000여 채가 부서지거나 침수되고 농경지는",
"refined": "In southern Russia, the thermohaline circulation system has been maintained, thank you for reporting. In North Korea, 22 people were killed and 4 went missing due to concentrated heavy rainfall and flooding. According to an email sent by the International Red Cross Federation to the Voice of America, citing the North Korean Red Cross and the National Emergency Management Committee, 22 people died and 4 went missing due to severe flooding in residential areas. In North Korea, over 8,000 homes were destroyed or flooded due to the flooding, and farmland was also damaged.",
"summarized": "Heavy rainfall in North Korea caused a flood that resulted in 22 deaths and 4 missing persons. The flood also destroyed or submerged around 8,000 homes and damaged over 2,000 hectares of farmland.",
"translated": "In North Korea, 22 people have died and 4 are missing due to floods caused by heavy rainfall. According to an email sent by the International Red Cross Federation to the Voice of America, the North Korean Red Cross Society and the National Emergency Management Committee reported that 22 people were killed and 4 are missing due to flooding in residential areas. Additionally, over 8,000 homes were destroyed or flooded, and a significant amount of farmland was damaged.",
"translated_llm": "In South Russia, we are keeping a fire spreading system. Thank you for your help. In North Korea, 22 people are hiding and 4 people are missing from the flood caused by concentrated flooding. The International Cross-Cross Federation in an email sent to the American Voice broadcast, according to the North Korean Cross-Cross-Country Conference and the National Civil Disaster Response Committee, citizens have been told that 22 people have died and 4 people are missing from the flood caused by the flood.",
"translated_whisper": " As a result of a concentration bomb in North Korea, 22 people were killed and 4 went missing. According to a email sent by the International Assembling Federation to Sori TV and U.S. At the time, NKWD and the National Institute for Emergency Operations reported 22 people dead or missing from a bomb explosion. The association said eight thousand homes have been attacked or taken over. Over 2,000 hectares of the farmland ended up being destroyed."
} |
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 citi... | [
"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 flo... | [
"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 duri... | {
"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"
} | 32 | [
"The news anchor is standing in front of a screen that shows a building with the sign \"K water.\" He is reporting on a situation where the Korea Water Resources Corporation is discussing the management of dam water levels during a recent flood. The anchor mentions that there was no issue with controlling the water... | nR08Eart4vo | A news report discusses the recent flooding incident involving the Korea Water Resources Corporation's management of dam water levels. Despite the corporation's claims that the water level was controlled according to regulations, considering weather changes, safety, and civilian concerns, the unprecedented prolonged rainy season led to the collapse of the dam and subsequent flooding. The report highlights the extensive damage caused by the flood, with aerial views showing submerged fields, houses, and infrastructure. The investigation into the causes of the flood is ongoing, with the prolonged rainy season identified as a significant contributing factor. The video also shows the aftermath of the flood, including damaged greenhouses and workers cleaning up the remains. The Korea Water Resources Corporation's explanation and the controversy over dam usage and flood control measures are also discussed. Overall, the video depicts the severe impact of the flood on the local community and the challenges faced in managing and recovering from the disaster. | {
"original": " 땜 관리를 담당하는 한국수자원공사가 이번 수해와 관련해 땜 수위 조절에는 문제가 없었고 규정에 따라 홍수기 기상 변화와 땜 상하류 상황 땜 안전과 민원 등을 고려해 홍수 대응이 가능하도록 땜 수위를 관리했다고 설명했습니다. 다만 유례없는 최장기간의 장마로 재방이 무너지고 물이 넘치는 등 복합요인 탓에 피해가 난 것으로 보인다고 덧붙였습니다. 감사합니다.",
"refined": "The South Korea Water Resources Administration, in charge of dam management, stated that there were no issues with controlling the water level, and the water level was managed according to regulations, considering flood season weather patterns, upstream and downstream conditions, dam safety, and civilian complaints to respond to flooding. However, it appears that the damage was caused by a combination of factors, including an unprecedented prolonged rainy season, which led to erosion and overflow.",
"summarized": "The Korea Water Resources Administration stated that there were no issues with controlling the water level during the recent flooding and that they managed the water level in accordance with regulations, considering factors such as flood season weather changes, upstream and downstream conditions, dam safety, and civilian concerns. However, the administration believes that the unprecedented long-duration rain caused the embankment to collapse, resulting in flooding and damage due to a combination of factors.",
"translated": "The Korea Water Resources Corporation, which is in charge of managing the dam, explained that there was no issue with controlling the water level during the recent flood, and that the water level was managed in accordance with regulations, taking into account changes in flood-related weather and the situation upstream and downstream of the dam, as well as safety and civilian concerns, to ensure a response to flooding. However, it added that the damage appeared to have been caused by a combination of factors, including the unprecedented prolonged rainy season, which led to the collapse of the dam and subsequent flooding.",
"translated_llm": "The Korean Water Authority, which is responsible for the management of the flood, said that there was no problem with the water control and that the water control was done in accordance with the regulations, taking into account the flood weather changes and the upstream situation, safety and civil society, etc.",
"translated_whisper": " South Korea Water Resources Administration handling the water system said it had no problem controlling the water level and the regulation requires the water levels changed and the water level level of the dam, and the safety, water quality status and civilian living in the dam should be considered in order to respond to flood damage. But, it appeared that the dam was damaged on the part of a flood which is harmful for our river."
} |
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 citi... | [
"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 flo... | [
"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 duri... | {
"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"
} | 32 | [
"In the current frame, two news anchors are seated in a studio, delivering a weather report. The screen behind them shows multiple monitors displaying various images, likely related to weather conditions. The text on the screen indicates a focus on heavy rainfall in the central region and the lifting of flood warni... | GwY9HYVEQIk | The video reports on the aftermath of heavy rainfall in the central region, which has caused significant damage to infrastructure, particularly water supply facilities. The heavy rainfall led to flood warnings being issued for major rivers nationwide, including the Nakdong River, where Samnangjin Bridge is located. However, as the rainfall subsided, flood warnings were gradually lifted, with the warning for Samnangjin Bridge being the last to be lifted. The damage from the heavy rainfall was extensive, with 93 water supply facilities in 27 cities and counties affected. The video shows footage of flooded rivers, damaged bridges, and water supply facilities, highlighting the challenges in recovery and infrastructure repair. The report also mentions that the country is experiencing its longest-ever monsoon season, which has set a new record, further emphasizing the prolonged impact of the heavy rainfall on the region's infrastructure and water supply systems. Overall, the video provides an update on the flood situation and the ongoing efforts to address the damage caused by the heavy rainfall. | {
"original": " 중부지방 집중호우가 소강상태에 접어든 가운데 전국 하천 주요 지점의 홍수특보가 전부 해제됐습니다. 환경부는 낙동강이 흐르는 경남 밀양시 삼랑진교에 내려졌던 홍수주의보가 오늘 오후 5시 30분 해제된 것을 끝으로 전국 하천 주요 지점에 내려진 홍수특보가 모두 해제됐다고 밝혔습니다. 한강과 임진강 비룡대교, 영평교, 왕숙천 진관교 등지에 발령됐던 홍수주의보와 섬진강 하동근 읍내리 지점의 홍수주의보는 각각 지난 11일 오후 6시와 오후 5시에 해제됐습니다. 이번 호우로 27개 시군의 상수도 시설 93곳에서 피해가 발생했습니다.",
"refined": "Heavy rainfall in the central region has subsided, and flood warnings have been lifted at major rivers nationwide. The Ministry of Environment announced that the flood warning issued for Samnangjin Bridge in Miryang, South Gyeongsang Province, where the Nakdong River flows, was lifted at 5:30 pm today. This marks the end of all flood warnings at major rivers across the country. Flood warnings previously issued for the Han River, Imjin River, and Seomjin River were lifted at 6 pm and 5 pm on the 11th. This round of heavy rainfall caused damage to 93 water treatment facilities in 27 cities and counties.",
"summarized": "Heavy rainfall in the central region has subsided, and flood warnings have been lifted from all major national river points. The Ministry of Environment reported that the flood warning for the Nakdong River in Miryang, Gyeongnam, was lifted at 5:30 pm, marking the end of all flood warnings nationwide. Other flood warnings, including those for the Han River and Seomjin River, were lifted on the 11th. The heavy rainfall caused damage to 93 water supply facilities in 27 cities and counties.",
"translated": "As the heavy rainfall in the central region has subsided, flood warnings have been lifted from all major rivers nationwide. The Ministry of Environment announced that the flood warning issued for Samnangjin Bridge in Miryang, South Gyeongsang Province, where the Nakdong River flows, was lifted at 5:30 pm today, marking the end of flood warnings for all major rivers. Previously, flood warnings had been lifted for the Han River, Imjin River, and Seomjin River, including the areas around Biyeong Bridge, Yeongpyeong Bridge, and Wangsukcheon Jinwan Bridge, at 6:00 pm and 5:00 pm on the 11th. This heavy rainfall caused damage to 93 water supply facilities in 27 cities and counties.",
"translated_llm": "The Central District Concentrated Flood Warnings have been cancelled. The Department of Environment has announced that flood warnings from the Central District Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the Central District flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood warnings from the district flood.",
"translated_whisper": " The Hongsu Special Agency has been removed from all the major national hot springs in the country while the Central Provincial Municipality of Gyeongsang Province is in a state of emergency. The Ministry of Environment said that the Hongsu Special Agency, which was dropped at Samnangjin Bridge in Miryang, Gyeongnam, where the Nakdong River flows, was removed at 5.30 pm today afternoon. It was also said that all the special services that had been dropped at major national hot springs in the country were removed. Hongsu and Seomjin River, which were located in the Hwadong-gun Eumnaeri area, were released at 6 p.m. and 5 p.m. on the 11th. This afternoon, damage occurred in 93 of the 27 damage occurred in 93 places."
} |
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 citi... | [
"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 flo... | [
"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 duri... | {
"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"
} | 32 | [
"In the current frame, a group of people is seen wearing face masks, suggesting a public health concern or response to an outbreak. The text in the image reads \"장기화된 코로나19 사태로\" and \"대한민국,\" indicating that the situation involves the prolonged COVID-19 pandemic in South Korea. The individuals appear to be in a cr... | FG82wp__uZ8 | The video depicts the impact of record-breaking rainfall on temples in South Korea, particularly highlighting the damage and flooding caused by the severe weather. The footage shows the initial awareness of the COVID-19 pandemic, with people wearing face masks, before transitioning to the effects of the heavy rainfall, which includes flooded temple grounds, damaged roofs, and destroyed infrastructure. News reporters are seen delivering updates on the worsening damage situation, emphasizing the need for urgent attention to preserve historical sites. The video also showcases the collective effort of monks and community members working together to clean up and restore the damaged temples, demonstrating their dedication and teamwork. Throughout the video, there are appeals for support and donations to aid the affected temples and communities, with contact information and bank account details provided for those willing to contribute. The overall context of the video is one of community resilience and the importance of collective effort in overcoming adversity, set against the backdrop of severe weather events and a global pandemic. | {
"original": " 기분 좋은 변화 품격 있는 강남 NEWS 기록적인 폭우가 이어지고 있는 가운데 전국 사찰 역시 피해 상황이 가중되고 있습니다. 한글자막 by 한효주",
"refined": "Good news of a pleasant change, the dignified Gangnam NEWS is reporting that historic heavy rains are continuing, and damage to temples nationwide is worsening.",
"summarized": "Not Available",
"translated": "A pleasant change, a dignified Gangnam NEWS: As the record-breaking rainfall continues, the damage situation is also worsening in temples across the country.",
"translated_llm": "The situation is also worsening in the national police as the record-breaking fire continues.",
"translated_whisper": " The United States has declared the situation of international"
} |
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 citi... | [
"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 flo... | [
"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 duri... | {
"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"
} | 32 | [
"A building with a blue roof is partially submerged in floodwaters. Several people are visible inside the structure, navigating through the water. The scene suggests a flood situation, possibly due to heavy rainfall. Trees and debris are scattered around, indicating the impact of the flood on the surrounding area."... | 1291976724456992776 | A severe flood has affected an area, partially submerging a building with a blue roof in water. The building, possibly a shelter or open structure, is crowded with people trying to manage the flooding situation. Debris, including pieces of wood and various colored objects, is scattered in the water, indicating the severity of the flood. Surrounding trees are partially submerged, and the water flows vigorously around them and the building. An urgent search and rescue operation is underway, as suggested by the audio transcript, with efforts to locate something or someone still in progress. The situation persists throughout the video, with the floodwaters continuing to affect the area and the search operation ongoing. Animals, such as water buffalo, are also seen struggling in the floodwaters near the building. The overall context is one of crisis and emergency response, with the flooding causing significant disruption and danger to people and animals in the affected area. | {
"original": " Baik, aku keliar nisik yaa, iya! Kami paham makhlam, kita dapat udi akta, di mana ya!",
"refined": "Not Generated",
"summarized": "Not Generated",
"translated": "Okay, I'm looking for you, yeah! We understand, we've got it, where is it?",
"translated_llm": "Okay, I'm looking for you, yeah! We understand Makhlam, we got you a deed, di mana ya!",
"translated_whisper": " It's sunny at home now. It looks like my friend who is looking at me will come back. This baby bird has little feathers in the back. He's so cute, isn't he?"
} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.