paper_id
stringlengths 10
10
⌀ | yes_no
bool 2
classes | paper_index
int64 0
519
⌀ | evidence
stringlengths 0
37.7k
| question
stringlengths 4
11.7k
| answer
stringlengths 1
26k
|
|---|---|---|---|---|---|
null | false
| 163
|
The database consists of 1,000 short texts, each having 10-40 words. Out of these, 350 are sarcastic and are collected as follows: (a) 103 sentences are from two popular sarcastic quote websites, (b) 76 sarcastic short movie reviews are manually extracted from the Amazon Movie Corpus BIBREF11 by two linguists. (c) 171 tweets are downloaded using the hashtag #sarcasm from Twitter. The 650 non-sarcastic texts are either downloaded from Twitter or extracted from the Amazon Movie Review corpus. The sentences do not contain words/phrases that are highly topic or culture specific. The tweets were normalized to make them linguistically well formed to avoid difficulty in interpreting social media lingo. Every sentence in our dataset carries positive or negative opinion about specific “aspects”. For example, the sentence “The movie is extremely well cast” has positive sentiment about the aspect “cast”.
The annotators were seven graduate students with science and engineering background, and possess good English proficiency. They were given a set of instructions beforehand and are advised to seek clarifications before they proceed. The instructions mention the nature of the task, annotation input method, and necessity of head movement minimization during the experiment.
The database consists of 1,000 short texts, each having 10-40 words.
|
How many short texts are analyzed in the study?
|
1000 short texts.
|
null | false
| null |
When did Alaska become a state?
|
Alaska became a state on January 3, 1959.
|
|
null | false
| null |
"Nothing" is the first single by English rock band A released from their album Hi-Fi Serious. It reached number nine in the UK Singles Chart; to date, A's highest charting single. During the promotion of the single, the band appeared on Top of The Pops and the Pepsi Chart Show.
The promotional video for Nothing was filmed in Cape Town, South Africa, mostly around the Artscape Theatre Centre formerly known as the Nico Malan theatre complex, and features extras dressed in the same clothes as each band member - complete with band member masks - to create the illusion that there are hundreds of duplicates of the band.
"Nothing" is arguably the heaviest A single, marking something of a departure from the band's usual melodic pop punk to nu metal. It is track 1 on the album Hi-Fi Serious. A live version appears on the live album, Rockin' Like Dokken.
|
Where was the heaviest A single filmed?
|
A's first hit single Nothing was filmed in Cape Town, South Africa.
|
null | false
| null |
In the UK, the recommended daily values for magnesium are 300 mg for men and 270 mg for women. In the U.S. the Recommended Dietary Allowances (RDAs) are 400 mg for men ages 19–30 and 420 mg for older; for women 310 mg for ages 19–30 and 320 mg for older.
|
Given this reference test, what is the daily recommended value of magnesium for men in the U.S.?
|
The daily recommended value of magnesium for men in the U.S. is 400 mg.
|
null | false
| null |
What are the different ways I can open a jar whose lid is stuck and will not open?
|
1. Using a rubber band or rubber glove - Surround the circumference of the lid with a rubber band and now try to open the lid the same way you would normally do. The rubber band creates additional friction which causes helps open the lid with less force. The same result can be achieved by using rubber glove.
2. Using hot water - Hold the jar under hot water, the hot water will allow the metal lid to expand allowing you to open the jar.
3. Using a knife - Break the vacuum seal by poking a hole in the lid using a sharp object such as a knife. Once the vacuum seal is broken, you can twist open the lid with less force.
|
|
null | false
| 28
|
We briefly review the task of generating natural language descriptions of semi-structured data, which we refer to as tables henceforth BIBREF11 , BIBREF12 . Tables can be expressed as set of records INLINEFORM0 , where each record is a tuple (entity, attribute, value). When all the records are about the same entity, we can truncate the records to (attribute, value) pairs. For example, for the table in Figure FIGREF2 , the records are {(Birth Name, Michael Dahlquist), (Born, December 22 1965), ...}. The task is to generate a text INLINEFORM1 which summarizes the records in a fluent and grammatical manner. For training and evaluation we further assume that we have a reference description INLINEFORM2 available for each table. We let INLINEFORM3 denote an evaluation set of tables, references and texts generated from a model INLINEFORM4 , and INLINEFORM5 , INLINEFORM6 denote the collection of n-grams of order INLINEFORM7 in INLINEFORM8 and INLINEFORM9 , respectively. We use INLINEFORM10 to denote the count of n-gram INLINEFORM11 in INLINEFORM12 , and INLINEFORM13 to denote the minimum of its counts in INLINEFORM14 and INLINEFORM15 . Our goal is to assign a score to the model, which correlates highly with human judgments of the quality of that model.
In Figure 1, the reference, though technically correct and fluent, mentions information which cannot be gleaned from the associated table. It also fails to mention useful information which a generation system might correctly include (e.g. candidate 3 in the figure). We call such references divergent from the associated table.
|
In the example of Figure 1, why do the team consider the reference text divergent from the associated table?
|
Because it mentions information which cannot be gleaned from the associated table. It also fails to mention useful information which a generation system might correctly include.
|
1805.02400
| false
| null |
We use the Yelp Challenge dataset BIBREF2 for our fake review generation. The dataset (Aug 2017) contains 2.9 million 1 –5 star restaurant reviews. We treat all reviews as genuine human-written reviews for the purpose of this work, since wide-scale deployment of machine-generated review attacks are not yet reported (Sep 2017) BIBREF19 . As preprocessing, we remove non-printable (non-ASCII) characters and excessive white-space. We separate punctuation from words. We reserve 15,000 reviews for validation and 3,000 for testing, and the rest we use for training. NMT models require a parallel corpus of source and target sentences, i.e. a large set of (source, target)-pairs. We set up a parallel corpus by constructing (context, review)-pairs from the dataset. Next, we describe how we created our input context.
We use the Yelp Challenge dataset BIBREF2 for our fake review generation.
|
Which dataset do they use a starting point in generating fake reviews?
|
The answers are shown as follows:
* the Yelp Challenge dataset
|
null | false
| null |
World War I was fought on many fronts around the world from the battlefields of Europe to the far-flung colonies in the Pacific and Africa. While it is most famous for the trench warfare stalemate that existed on Europe's Western Front, in other theatres of combat the fighting was mobile and often involved set-piece battles and cavalry charges. The Eastern Front often took thousands of casualties a day during the major offensive pushes, but it was the West that saw the most concentrated slaughter. It was in the west that the newly industrialized world powers could focus their end products on the military–industrial complex. The deadliest day of the war was during the opening days of the conflict. The Imperial German war council had initiated the Schlieffen Plan which involved multiple armies flooding through the borders of Belgium and France. On August 22, 1914, during the Battle of the Frontiers, five separate French armies engaged the German invaders independently of each other. Across all those battlefields, on that single day, 27,000 French soldiers lost their lives protecting their country.
|
Please describe what combat was like in World War I
|
World War 1 was fought in Europe, the Pacific, and Africa. It involved cavalry, trench warfare, and mechanized combat. While World War I is known primarily for trench warfare, it also involved mobile combat and the movement of tens of thousands of men, machines, and animals. There were numerous, bloody days and battles in World War I, but the most deadly took place during the opening days of the conflict, at the Battle of Frontiers, where 27,000 French soldiers were killed in a single day.
|
2004.03744
| false
| null |
e-SNLI-VE-2.0 is the combination of SNLI-VE-2.0 with explanations from either e-SNLI or our crowdsourced annotations where applicable. The statistics of e-SNLI-VE-2.0 are shown in Table TABREF40.
FLOAT SELECTED: Table 3. Summary of e-SNLI-VE-2.0 (= SNLI-VE-2.0 + explanations). Image-sentence pairs labelled as neutral in the training set have not been corrected.
The statistics of e-SNLI-VE-2.0 are shown in Table TABREF40.
FLOAT SELECTED: Table 3. Summary of e-SNLI-VE-2.0 (= SNLI-VE-2.0 + explanations). Image-sentence pairs labelled as neutral in the training set have not been corrected.
|
How many natural language explanations are human-written?
|
Totally 6980 validation and test image-sentence pairs have been corrected.
|
null | false
| 33
|
This section describes the similarity and relatedness measures we integrate in our second–order co–occurrence vectors. We use two taxonomies in this study, SNOMED–CT and MeSH. SNOMED–CT (Systematized Nomenclature of Medicine Clinical Terms) is a comprehensive clinical terminology created for the electronic representation of clinical health information. MeSH (Medical Subject Headings) is a taxonomy of biomedical terms developed for indexing biomedical journal articles.
We obtain SNOMED–CT and MeSH via the Unified Medical Language System (UMLS) Metathesaurus (version 2016AA). The Metathesaurus contains approximately 2 million biomedical and clinical concepts from over 150 different terminologies that have been semi–automatically integrated into a single source. Concepts in the Metathesaurus are connected largely by two types of hierarchical relations: INLINEFORM0 / INLINEFORM1 (PAR/CHD) and INLINEFORM2 / INLINEFORM3 (RB/RN).
We use two taxonomies in this study, SNOMED–CT and MeSH.
|
What are the two taxonomies used in this study?
|
SNOMED–CT and MeSH.
|
1910.05154
| false
| null |
The experiment settings from this paper and evaluation protocol for the Mboshi corpus (Boundary F-scores using the ZRC speech reference) are the same from BIBREF8. Table presents the results for bilingual UWS and multilingual leveraging. For the former, we reach our best result by using as aligned information the French, the original aligned language for this dataset. Languages closely related to French (Spanish and Portuguese) ranked better, while our worst result used German. English also performs notably well in our experiments. We believe this is due to the statistics features of the resulting text. We observe in Table that the English portion of the dataset contains the smallest vocabulary among all languages. Since we train our systems in very low-resource settings, vocabulary-related features can impact greatly the system's capacity to language-model, and consequently the final quality of the produced alignments. Even in high-resource settings, it was already attested that some languages are more difficult to model than others BIBREF9.
For the multilingual selection experiments, we experimented combining the languages from top to bottom as they appear Table (ranked by performance; e.g. 1-3 means the combination of FR(1), EN(2) and PT(3)). We observe that the performance improvement is smaller than the one observed in previous work BIBREF10, which we attribute to the fact that our dataset was artificially augmented. This could result in the available multilingual form of supervision not being as rich as in a manually generated dataset. Finally, the best boundary segmentation result is obtained by performing multilingual voting with all the languages and an agreement of 50%, which indicates that the information learned by different languages will provide additional complementary evidence.
Lastly, following the methodology from BIBREF8, we extract the most confident alignments (in terms of ANE) discovered by the bilingual models. Table presents the top 10 most confident (discovered type, translation) pairs. Looking at the pairs the bilingual models are most confident about, we observe there are some types discovered by all the bilingual models (e.g. Mboshi word itua, and the concatenation oboá+ngá). However, the models still differ for most of their alignments in the table. This hints that while a portion of the lexicon might be captured independently of the language used, other structures might be more dependent of the chosen language. On this note, BIBREF11 suggests the notion of word cannot always be meaningfully defined cross-linguistically.
The experiment settings from this paper and evaluation protocol for the Mboshi corpus (Boundary F-scores using the ZRC speech reference) are the same from BIBREF8. Table presents the results for bilingual UWS and multilingual leveraging. For the former, we reach our best result by using as aligned information the French, the original aligned language for this dataset.
For the multilingual selection experiments, we experimented combining the languages from top to bottom as they appear Table (ranked by performance; e.g. 1-3 means the combination of FR(1), EN(2) and PT(3)). We observe that the performance improvement is smaller than the one observed in previous work BIBREF10, which we attribute to the fact that our dataset was artificially augmented.
Lastly, following the methodology from BIBREF8, we extract the most confident alignments (in terms of ANE) discovered by the bilingual models. Table presents the top 10 most confident (discovered type, translation) pairs. Looking at the pairs the bilingual models are most confident about, we observe there are some types discovered by all the bilingual models (e.g. Mboshi word itua, and the concatenation oboá+ngá).
|
How is the performance of the model evaluated?
|
The answers are shown as follows:
* The experiment settings from this paper and evaluation protocol for the Mboshi corpus (Boundary F-scores using the ZRC speech reference) are the same from BIBREF8.
* For the multilingual selection experiments, we experimented combining the languages from top to bottom as they appear Table (ranked by performance; e.g. 1-3 means the combination of FR(1), EN(2) and PT(3)).
* Lastly, following the methodology from BIBREF8, we extract the most confident alignments (in terms of ANE) discovered by the bilingual models.
|
null | false
| null |
What is a tennis leg injury?
|
A tennis leg injury is an incomplete rupture of the inside of the calf muscle, where multiple muscles are attached. It is caused mostly because of tennis-like sports due to a sudden contraction of the calf muscles. Sports like tennis and squash are most likely causing this kind of trouble due to the short and intense moves it requires to practice.
|
|
null | false
| null |
First day of baking with sourdough
|
Today is my first day baking with sourdough. The process seems very time consuming as you first need to create a sourdough starter. Creating the starter takes usually seven days.
Following baking instructions can also be challenging as you need to be very careful and make sure that all ingredients are added in a certain order.
There are many different kinds of flour, all with their own baking characteristics, making the process even more difficult.
I am nervous about this, but I've always wanted to learn this craft as baking with sourdough has been done since the beginning of humanity.
|
|
null | false
| null |
What is the new record of Javelin throw made by Neeraj Chopra
|
89.94 m
|
|
null | false
| null |
Identify which instrument is string or percussion: Rakatak, K'ni
|
K'ni is string, Rakatak is percussion.
|
|
null | false
| 396
|
Our political systems are unequal, and we suffer for it. Diversity in representation around decision-making tables is important for the health of our democratic institutions BIBREF0. One example of this inequity of representation is the gender disparity in politics: there are fewer women in politics than men, largely because women do not run for office at the same rate as men. This is because women face systemic barriers in political systems across the world BIBREF1. One of these barriers is online harassment BIBREF2, BIBREF3. Twitter is an important social media platform for politicians to share their visions and engage with their constituents. Women are disproportionately harassed on this platform because of their gender BIBREF4.
To raise awareness of online abuse and shift the discourse surrounding women in politics, we designed, built, and deployed ParityBOT: a Twitter bot that classifies hateful tweets directed at women in politics and then posts “positivitweets”. This paper focuses on how ParityBOT improves discourse in politics.
Previous work that addressed online harassment focused on collecting tweets directed at women engaged in politics and journalism and determining if they were problematic or abusive BIBREF5, BIBREF3, BIBREF6. Inspired by these projects, we go one step further and develop a tool that directly engages in the discourse on Twitter in political communities. Our hypothesis is that by seeing “positivitweets” from ParityBOT in their Twitter feeds, knowing that each tweet is an anonymous response to a hateful tweet, women in politics will feel encouraged and included in digital political communitiesBIBREF7. This will reduce the barrier to fair engagement on Twitter for women in politics. It will also help achieve gender balance in Canadian politics and improve gender equality in our society.
To raise awareness of online abuse and shift the discourse surrounding women in politics, we designed, built, and deployed ParityBOT: a Twitter bot that classifies hateful tweets directed at women in politics and then posts “positivitweets”.
|
What is ParityBOT?
|
A Twitter bot that classifies hateful tweets directed at women in politics.
|
null | false
| null |
Classify the below car manufacturing companies based on their countries of origin.
Mazda, KIA, Hyundai, Ford, Honda, Toyota, Chevrolet
|
Japan - Toyota, Mazda
Korea - South Korea
USA - Chevrolet, Ford
|
|
null | false
| 487
|
As we discussed in Section 3, GNP generalizes sum, max, mean, and min. In order to confirm the facts experimentally, we showed in Figure how the learnable parameters p and q in GNP changed during training. For maxdegree, GNP as aggregation and readout functions approximated sum (i.e., p + ≈ 1 and q + ≈ 0) and max (i.e., p + 1 and q + ≈ 0), respectively, which performed best on the task. For µ post and shortest, GNP approximated mean (i.e., p + ≈ 1 and q + ≈ 1) and min (i.e., p − 0 and q − ≈ 0), respectively, which were the best performing baseline for the tasks. To sum up, empirically, GNP imitated the ideal pooling functions for each task if such pooling functions exist. We also observed that either GNP + or GNP − tends to dominate the other side in all considered graph-level tasks. Detailed results are provided in Appendix C.6.
As we discussed in Section 3, GNP generalizes sum, max, mean, and min. In order to confirm the facts experimentally, we showed in Figure how the learnable parameters p and q in GNP changed during training. For maxdegree, GNP as aggregation and readout functions approximated sum (i.e., p + ≈ 1 and q + ≈ 0) and max (i.e., p + 1 and q + ≈ 0), respectively, which performed best on the task. For µ post and shortest, GNP approximated mean (i.e., p + ≈ 1 and q + ≈ 1) and min (i.e., p − 0 and q − ≈ 0), respectively, which were the best performing baseline for the tasks. To sum up, empirically, GNP imitated the ideal pooling functions for each task if such pooling functions exist. We also observed that either GNP + or GNP − tends to dominate the other side in all considered graph-level tasks. Detailed results are provided in Appendix C.6.
As we discussed in Section 3, GNP generalizes sum, max, mean, and min. In order to confirm the facts experimentally, we showed in Figure 2 how the learnable parameters p and q in GNP changed during training. For maxdegree, GNP as aggregation and readout functions approximated sum (i.e., p + ≈ 1 and q + ≈ 0) and max (i.e., p + _x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f_x005f 1 and q + ≈ 0), respectively, which performed best on the task. For µpost and shortest, GNP approximated mean (i.e., p + ≈ 1 and q + ≈ 1) and min (i.e., p − _x001d_ 0 and q − ≈ 0), respectively, which were the best performing baseline for the tasks. To sum up, empirically, GNP imitated the ideal pooling functions for each task if such pooling functions exist. We also observed that either GNP+ or GNP− tends to dominate the other side in all considered graph-level tasks. Detailed results are provided in Appendix C.6.****Figure 2: Empirical behavior of GNP. We showed how the parameters p and q of GNP changed during training. For each task, GNP imitated the ideal pooling functions if such pooling functions exist. For example, for maxdegree, GNP as aggregation and readout functions approximated sum (i.e., p + ≈ 1 and q + ≈ 0) and max (i.e., p + 1 and q + ≈ 0), respectively, which performed best.
|
In case of min/max, what are the values of "p" learned by GNP?
|
We demonstrated the empirical behavior of the values of "p" learned by GNP in Section 4.5. For details, please refer to Figure 2 and Section 4.5 (in the revised paper).
|
null | false
| null |
Aries (♈︎) (Greek: Κριός, romanized: Kriós, Latin for "ram") is the first astrological sign in the zodiac, spanning the first 30 degrees of celestial longitude (0°≤ λ <30°), and originates from the Aries constellation. Under the tropical zodiac, the Sun transits this sign from approximately March 21 to April 19 each year. This time duration is exactly the first month of the Solar Hijri calendar (Arabic Hamal/Persian Farvardin/Pashto Wray).
|
For the Zodiac sign Aries, Share some information from the given text.
|
1. Aries is the first astrological sign in the zodiac, spanned in the first 30 degrees of celestial longitude (0°≤ λ <30°).
2. Aries is originated from the Aries constellation.
3. The Sun transits this sign from approximately March 21 to April 19 each year.
4. This time period is exactly the first month of the Solar Hijri calendar (Arabic Hamal/Persian Farvardin/Pashto Wray).
|
null | false
| 95
|
People are increasingly using social networking platforms such as Twitter, Facebook, YouTube, etc. to communicate their opinions and share information. Although the interactions among users on these platforms can lead to constructive conversations, they have been increasingly exploited for the propagation of abusive language and the organization of hate-based activities BIBREF0, BIBREF1, especially due to the mobility and anonymous environment of these online platforms. Violence attributed to online hate speech has increased worldwide. For example, in the UK, there has been a significant increase in hate speech towards the immigrant and Muslim communities following the UK's leaving the EU and the Manchester and London attacks. The US also has been a marked increase in hate speech and related crime following the Trump election. Therefore, governments and social network platforms confronting the trend must have tools to detect aggressive behavior in general, and hate speech in particular, as these forms of online aggression not only poison the social climate of the online communities that experience it, but can also provoke physical violence and serious harm BIBREF1.
Recently, the problem of online abusive detection has attracted scientific attention. Proof of this is the creation of the third Workshop on Abusive Language Online or Kaggle’s Toxic Comment Classification Challenge that gathered 4,551 teams in 2018 to detect different types of toxicities (threats, obscenity, etc.). In the scope of this work, we mainly focus on the term hate speech as abusive content in social media, since it can be considered a broad umbrella term for numerous kinds of insulting user-generated content. Hate speech is commonly defined as any communication criticizing a person or a group based on some characteristics such as gender, sexual orientation, nationality, religion, race, etc. Hate speech detection is not a stable or simple target because misclassification of regular conversation as hate speech can severely affect users’ freedom of expression and reputation, while misclassification of hateful conversations as unproblematic would maintain the status of online communities as unsafe environments BIBREF2.
To detect online hate speech, a large number of scientific studies have been dedicated by using Natural Language Processing (NLP) in combination with Machine Learning (ML) and Deep Learning (DL) methods BIBREF3, BIBREF4, BIBREF5, BIBREF6, BIBREF7, BIBREF0. Although supervised machine learning-based approaches have used different text mining-based features such as surface features, sentiment analysis, lexical resources, linguistic features, knowledge-based features or user-based and platform-based metadata BIBREF8, BIBREF9, BIBREF10, they necessitate a well-defined feature extraction approach. The trend now seems to be changing direction, with deep learning models being used for both feature extraction and the training of classifiers. These newer models are applying deep learning approaches such as Convolutional Neural Networks (CNNs), Long Short-Term Memory Networks (LSTMs), etc.BIBREF6, BIBREF0 to enhance the performance of hate speech detection models, however, they still suffer from lack of labelled data or inability to improve generalization property.
Here, we propose a transfer learning approach for hate speech understanding using a combination of the unsupervised pre-trained model BERT BIBREF11 and some new supervised fine-tuning strategies. As far as we know, it is the first time that such exhaustive fine-tuning strategies are proposed along with a generative pre-trained language model to transfer learning to low-resource hate speech languages and improve performance of the task. In summary:
We propose a transfer learning approach using the pre-trained language model BERT learned on English Wikipedia and BookCorpus to enhance hate speech detection on publicly available benchmark datasets. Toward that end, for the first time, we introduce new fine-tuning strategies to examine the effect of different embedding layers of BERT in hate speech detection.
Our experiment results show that using the pre-trained BERT model and fine-tuning it on the downstream task by leveraging syntactical and contextual information of all BERT's transformers outperforms previous works in terms of precision, recall, and F1-score. Furthermore, examining the results shows the ability of our model to detect some biases in the process of collecting or annotating datasets. It can be a valuable clue in using pre-trained BERT model for debiasing hate speech datasets in future studies.
In the scope of this work, we mainly focus on the term hate speech as abusive content in social media, since it can be considered a broad umbrella term for numerous kinds of insulting user-generated content.
|
Why do the authors focus on the term hate speech as abusive content in social media?
|
Since it can be considered a broad umbrella term for numerous kinds of insulting user-generated content.
|
null | false
| null |
Why is Canada a good country to visit?
|
Canada is a beautiful country known for its natural scenery, ice hockey, polite people, maple syrup and wildlife.
|
|
null | false
| null |
Identify which instrument is string or percussion: Paila criolla, Topshur
|
Topshur is string, Paila criolla is percussion.
|
|
null | false
| null |
Which is a species of fish? Trahira or Tahini
|
Trahira
|
|
null | false
| null |
What is an architecture diagram?
|
An architecture diagram is a visual layout that represents process flows, business process management workflows, technology workflows, data workflows, and more. They are a represented with different shapes, words, lines are arrows that show the visual flow. They help engineer, architect, stakeholders, and designers understand the layout of a project, current system design flaws, and how to make future state architecture more robust and resilient to change.
|
|
null | false
| null |
What was the SS Frosta
|
SS Frosta was a Norwegian oil tanker, built in 1961 in Germany by Bremer Vulcan and owned by A/S J. Ludwig Mowinckels Rederi of Bergen, Norway. The Frosta was 664 feet in length, 90 feet in breadth, with a gross weight of 22,850 tons, and powered by a steam turbine engine, rated at 16,800 horsepower. It was rebuilt as a chemical tanker in 1971. It was decommissioned in 1979.
|
|
null | false
| null |
What shapes are made up of straight lines?
|
Square, pentagon, rectangle, rhombus and triangle are examples of shapes made up only of straight lines.
|
|
null | false
| null |
What is the political system of Germany?
|
The Federal Republic of Germany is a federal state and is organized as a parliamentary democracy. The Basic Law stipulates that all state power emanates from the people. The people delegate this power to the parliaments (Bundestag and Landtage) for the duration of an election period.
State power is divided between the legislative branch, the executive branch and the power responsible for the administration of justice. This division of powers is an integral part of every democracy and in Germany it is enshrined in the Basic Law, the constitution. The legislative branch includes parliaments, the executive branch the chancellor and the ministers. The judiciary plays a central role because judges in the state and federal courts are independent and rule solely on the basis of the law. The highest German court is the Federal Constitutional Court , which monitors compliance with the Basic Law. All other state organs are bound by the case law of the 16 constitutional judges.
|
|
null | false
| null |
Should I have Kids?
|
That is a personal question and one should think about this very carefully. Things to consider:
1. How old are you?
2. Can you afford to raise the child
3. Do you have proper healthcare
4. Do you have anyone to help you
5. Do you like kids
6. Why do you want kids
|
|
null | false
| null |
Who are some of the most influential women's hockey players
|
Hayley Wickenheiser, Cassie Campbell, Cammi Granato, Angela Ruggiero
|
|
null | false
| null |
Thierry Daniel Henry (French pronunciation: [tjɛʁi danjɛl ɑ̃ʁi]; born 17 August 1977) is a French professional football coach, pundit, and former player. Considered one of the best strikers of all time, one of the best players to play in the Premier League and Arsenal F.C's greatest player of all time. Henry was runner-up for both the Ballon d'Or in 2003 and the FIFA World Player of the Year in 2003 and 2004. He was named the FWA Footballer of the Year a record three times, the PFA Players' Player of the Year a joint-record two times, and was named in the PFA Team of the Year six consecutive times. He was also included in the FIFA FIFPro World XI once and the UEFA Team of the Year five times.
Henry made his professional debut with Monaco in 1994 before signing for defending Serie A champions Juventus. However, limited playing time, coupled with disagreements with the club's hierarchy, led to him signing for Premier League club Arsenal for £11 million in 1999. Under long-time mentor and coach Arsène Wenger, Henry became a prolific striker and Arsenal's all-time leading scorer with 228 goals in all competitions. He won the Premier League Golden Boot a record four times, won two FA Cups and two Premier League titles with the club, including one during an unbeaten Invincible season. Henry spent his final two seasons with Arsenal as club captain, leading them to the 2006 UEFA Champions League Final. Henry transferred to Barcelona in 2007 and in the 2008–09 season, he was a key part of the club's historic treble when they won La Liga, the Copa del Rey, and the UEFA Champions League. In 2010, he joined Major League Soccer (MLS) club New York Red Bulls and returned to Arsenal on loan for two months in 2012, before retiring in 2014.
Henry had success with France, winning the 1998 FIFA World Cup, UEFA Euro 2000, and 2003 FIFA Confederations Cup. He was named French Player of the Year a record five times, named to the UEFA Euro 2000 Team of the Tournament, awarded both the 2003 FIFA Confederations Cup Golden Ball and Golden Shoe, and named to the 2006 FIFA World Cup All-Star Team. In October 2007, he became his country's record goalscorer, a record he held until December 2022. After amassing 123 appearances and 51 goals, Henry retired from international football after the 2010 FIFA World Cup.
|
Which french football player is considered one of the best strikers of all time?
|
Thierry Henry, born in 1977, who played professionally between 1994 and 2012, is considered one of the best strikers of all time. He won the FIFA World Cup with France in 1998, the European Cup in 2000 and the Confederations Cup in 2003. Henry turned coach after ending his professional career in 2012.
|
null | false
| 238
|
The ability to make effective presentations has been found to be linked with success at school and in the workplace. Humor plays an important role in successful public speaking, e.g., helping to reduce public speaking anxiety often regarded as the most prevalent type of social phobia, generating shared amusement to boost persuasive power, and serving as a means to attract attention and reduce tension BIBREF0 .
Automatically simulating an audience's reactions to humor will not only be useful for presentation training, but also improve conversational systems by giving machines more empathetic power. The present study reports our efforts in recognizing utterances that cause laughter in presentations. These include building a corpus from TED talks and using Convolutional Neural Networks (CNNs) in the recognition.
The remainder of the paper is organized as follows: Section SECREF2 briefly reviews the previous related research; Section SECREF3 describes the corpus we collected from TED talks; Section SECREF4 describes the text classification methods; Section SECREF5 reports on our experiments; finally, Section SECREF6 discusses the findings of our study and plans for future work.
These include building a corpus from TED talks and using Convolutional Neural Networks (CNNs) in the recognition.
|
Where did they collect the corpus?
|
TED talks.
|
null | false
| null |
In Islam, cateogorize each of the following as either 'haram' or 'halal'. Pork, Beef, Chicken, Fish, Alcohol, fasting.
|
Pork and Alcohol are considered Haram while Beef, Chicken, Fish and Fasting are Halal
|
|
null | false
| null |
Choose the word which is different from the rest: potassium, silicon, zirconium, gallium, germanium
|
All except potassium are metals used in semiconductor devices
|
|
null | false
| null |
write a short poem about gratitude to my sister friends who lead Obon dancing with me every year
|
A small group of women gather with me in the small country church in California to keep Japanese Obon (folk dancing) tradition alive in the community. The festival is open to the public and includes brightly colored kimonos, lanterns and japanese food. We carry the tradition for the next generation - children are dressed in traditional dress. These 8 - 10 sister-women keep the tradition of honoring our ancestors, keeping the family tradition and provide dance leadership to hundreds of people every year at the annual Obon festival. They sacrifice their time to teach and lead these dances every year. They are my sisters
|
|
null | false
| 16
|
For the neural projector, we employ rectified networks as coupling function INLINEFORM0 following BIBREF16 . We use a rectified network with an input layer, one hidden layer, and linear output units, the number of hidden units is set to the same as the number of input units. The number of coupling layers are varied as 4, 8, 16 for both tasks. We optimize marginal data likelihood directly using Adam BIBREF22 . For both tasks in the fully unsupervised setting, we do not tune the hyper-parameters using supervised data.
Following existing literature, we train and test on the entire WSJ corpus (49208 sentences, 1M tokens). We use 45 tag clusters, the number of POS tags that appear in WSJ corpus.
|
What corpus do they use for training and testing?
|
WSJ corpus.
|
null | false
| null |
Theodore Scott Yoho (/ˈjoʊhoʊ/ YOH-hoh; born April 13, 1955) is an American politician, veterinarian, and businessman who served as the U.S. representative from Florida's 3rd congressional district from 2013 until 2021. He is a member of the Republican Party.
In the 2012 Republican primary election for the district, Yoho pulled a major upset against long-term incumbent Congressman Cliff Stearns, who had first been elected in 1988. Prior to being elected to Congress, Yoho had been a veterinarian and small business owner in North Central Florida for 30 years.
In December 2019, Yoho announced that, honoring his term-limit pledge, he would not run for re-election to Congress in 2020.
|
Is Ted Yoho republican or democrat
|
Theodore Scott Yoho is an American politician, veterinarian, and businessman who served as the U.S. representative from Florida's 3rd congressional district from 2013 until 2021. He is a member of the Republican Party.
In the 2012 Republican primary election for the district, Yoho pulled a major upset against long-term incumbent Congressman Cliff Stearns, who had first been elected in 1988. Prior to being elected to Congress, Yoho had been a veterinarian and small business owner in North Central Florida for 30 years.
In December 2019, Yoho announced that, honoring his term-limit pledge, he would not run for re-election to Congress in 2020.
|
null | false
| null |
Captain Ginyu (ギニュー隊長, Ginyū-Taichō) is a horned purple-skinned humanoid who is the leader of the Ginyu Force. He is shown to be the most tactically skilled, as he deduced that Goku was masking his true power level by simply watching him fight. After seeing Goku's increased power level when using the Kaio-ken, Ginyu inflicts a near-fatal injury on himself,[ch. 285, 286] before using his technique that allows him to switch bodies with anyone, leaving Goku with Ginyu's nearly dead body.[ch. 286, 287] As Goku, Ginyu momentarily deceives Krillin but has trouble fighting Krillin and Gohan when he realizes that he can not fully control Goku's power.[ch. 288, 289] Ginyu then tries to switch into Vegeta's body, but Goku intervenes getting his body back, and when Ginyu tries again, he switches bodies with a Namekian frog Goku throws in the way.[ch. 290] In Dragon Ball Super, Ginyu is revealed to have survived Namek's destruction and escaped to Earth, though he is still trapped in the frog's body. He takes the opportunity to swap bodies with one of Frieza's soldiers named Tagoma, but is killed by Vegeta shortly afterward. Ginyu is voiced by Hideyuki Hori in the original series and Katsuyuki Konishi in Kai. In English, he is voiced by Richard Newman in the Ocean dub, Dale Kelly in Funimation's original dub, Brice Armstrong in their redub, and by Robert Bruce Elliott in Kai.
|
From this summary of Captain Ginyu's role in the series Dragon Ball, how many times did Captain Ginyu switch bodies?
|
Captain Ginyu switches bodies 4 times in the Dragon Ball series. Captain Ginyu first switched bodies with Goku and back. Then Captain Ginyu switched bodies a Namekian frog, then in Dragon Ball Super, he switched bodies with one of Frieza's soldiers named Tagoma, and is promptly killed by Vegeta.
|
null | false
| null |
Side A
1. "H2O Gate Blues" – 7:58 (about the Watergate break-in and cover-up) (from Winter in America)
2. "We Beg Your Pardon (Pardon our Analysis)" – 7:52 (about the pardon of Richard Nixon) (from The First Minute of a New Day)
3. "The New Deal" – 3:10
Side B
4. "Jose Campos Torres" – 2:36 (about Jose Campos Torres, a U.S. Army veteran who was arrested and then murdered and tossed into a bayou by two police officers in Houston in 1978, spurring the Moody Park Riot)
5. "The Ghetto Code (Dot Dot Dit Dit Dot Dot Dash)" – 12:57
6. "Bicentennial Blues" – 8:39 (from It's Your World)
|
What are the titles of the songs on Side B for this album?
|
"Jose Campos Torres"
"The Ghetto Code (Dot Dot Dit Dit Dot Dot Dash)"
"Bicentennial Blues"
|
null | false
| null |
What is the name of the dog that looks like a mop?
|
There are a number of dogs with mop-like appearances. The two most famous examples are the Komondor, which is a Hungarian guard dog, and the Puli, which was bred as a herding dog. Another example includes the lesser-known Bergamasco Sheepdog.
|
|
null | false
| null |
Name 5 possible fuel sources for a futuristic spaceship
|
Ion fusion, hyper drive, flux capacitator, solar sails, alien venom
|
|
null | false
| null |
Why should you exercise if you have a sitting job where you have to sit in front of computer screen ?
|
Basically exercise has many benefits listed below
1. Exercise manages weight
2. Exercise prevents and treats illnesses
3. Exercise enhances mood
4. Exercise gives you more energy
5. Exercise helps you sleep better.
6. Exercise revitalises your personal life.
|
|
null | false
| null |
Write a short story about an inn staffed by bears, birds, and sea creatures. The story should include details about the staff and the activities that the inn offers.
|
There once was a small inn in San Francisco called the Crab Inn. Its proprietor was a friendly and hard-working stone crab named Joe. Joe was an inspiring leader and had a knack for managing diverse species, thus he had attracted a sizable contingent of bears and birds to work at the inn. His most valued employees were a large polar bear who served as CFO, and a brown bear from Britain who was the hotel’s concierge. Additional staff members included a large brown bear who ran the shipping and receiving department, and a hen whose feathers made her an excellent housekeeper. Security was provided by a retired Canadian police bear and a retired British police bear. And because Joe recognized the importance of staying at the forefront of innovation, the inn employed a mole to research new hotel concepts. Lastly, the inn had a small polar bear on retainer as their attorney, and an Irish doctor bear to look after the medical needs of the staff and the very occasional guest who would stay at the inn. Amazingly, Joe and his CFO had convinced the staff to work in return only for housing and an occasional chocolate coin.
Although the inn was very small, it offered guests various activities. The most popular was a whale watching cruise. Although the ship’s captain, a penguin, was not a skilled navigator, guests were guaranteed whale sightings on every voyage because the whales worked for the inn. The inn’s spa was focused on a small gym facility, and an emotional support bear offered counseling services so that guests could heal both their bodies and minds. The most popular guest dining spot was al fresco in the inn’s award-winning garden, where a pair of geese served as bartender and wine sommelier. Lastly, guests enjoyed parties organized by the hotel’s panda bear to watch broadcasts of local sports teams on the inn’s large television.
|
|
null | false
| null |
Identify which instrument is string or percussion: Zabumba, Dan ghita
|
Dan ghita is string, Zabumba is percussion.
|
|
null | false
| null |
Easter eggs, also called Paschal eggs, are eggs that are decorated for the Christian feast of Easter, which celebrates the resurrection of Jesus. As such, Easter eggs are common during the season of Eastertide (Easter season). The oldest tradition, which continues to be used in Central and Eastern Europe, is to use dyed and painted chicken eggs.
Although eggs, in general, were a traditional symbol of fertility and rebirth, in Christianity, for the celebration of Eastertide, Easter eggs symbolize the empty tomb of Jesus, from which Jesus was resurrected. In addition, one ancient tradition was the staining of Easter eggs with the colour red "in memory of the blood of Christ, shed as at that time of his crucifixion."
This custom of the Easter egg, according to many sources, can be traced to early Christians of Mesopotamia, and from there it spread into Eastern Europe and Siberia through the Orthodox Churches, and later into Europe through the Catholic and Protestant Churches. Additionally, the widespread usage of Easter eggs, according to mediaevalist scholars, is due to the prohibition of eggs during Lent after which, on Easter, they have been blessed for the occasion.
A modern custom in some places is to substitute chocolate eggs wrapped in coloured foil, hand-carved wooden eggs, or plastic eggs filled with confectionery such as chocolate.
|
Given a reference text about Easter eggs, tell me what they are, what they symbolize, and where they originated from?
|
Easter eggs are decorated eggs celebrating the Christian holiday of Easter. They symbolize the empty tomb of Jesus. They originated in Mesopotamia.
|
null | false
| null |
Identify which instrument is string or percussion: Bell, Tro
|
Tro is string, Bell is percussion.
|
|
null | false
| 283
|
Scripts were developed as a means of representing stereotypical event sequences and interactions in narratives. The benefits of scripts for encoding common sense knowledge, filling in gaps in a story, resolving ambiguous references, and answering comprehension questions have been amply demonstrated in the early work in natural language understanding BIBREF0 . The earliest attempts to learn scripts were based on explanation-based learning, which can be characterized as example-guided deduction from first principles BIBREF1 , BIBREF2 . While this approach is successful in generalizing from a small number of examples, it requires a strong domain theory, which limits its applicability.
More recently, some new graph-based algorithms for inducing script-like structures from text have emerged. “Narrative Chains” is a narrative model similar to Scripts BIBREF3 . Each Narrative Chain is a directed graph indicating the most frequent temporal relationship between the events in the chain. Narrative Chains are learned by a novel application of pairwise mutual information and temporal relation learning. Another graph learning approach employs Multiple Sequence Alignment in conjunction with a semantic similarity function to cluster sequences of event descriptions into a directed graph BIBREF4 . More recently still, graphical models have been proposed for representing script-like knowledge, but these lack the temporal component that is central to this paper and to the early script work. These models instead focus on learning bags of related events BIBREF5 , BIBREF6 .
While the above approches demonstrate the learnability of script-like knowledge, they do not offer a probabilistic framework to reason robustly under uncertainty taking into account the temporal order of events. In this paper we present the first formal representation of scripts as Hidden Markov Models (HMMs), which support robust inference and effective learning algorithms. The states of the HMM correspond to event types in scripts, such as entering a restaurant or opening a door. Observations correspond to natural language sentences that describe the event instances that occur in the story, e.g., “John went to Starbucks. He came back after ten minutes.” The standard inference algorithms, such as the Forward-Backward algorithm, are able to answer questions about the hidden states given the observed sentences, for example, “What did John do in Starbucks?”
There are two complications that need to be dealt with to adapt HMMs to model narrative scripts. First, both the set of states, i.e., event types, and the set of observations are not pre-specified but are to be learned from data. We assume that the set of possible observations and the set of event types to be bounded but unknown. We employ the clustering algorithm proposed in BIBREF4 to reduce the natural language sentences, i.e., event descriptions, to a small set of observations and states based on their Wordnet similarity.
The second complication of narrative texts is that many events may be omitted either in the narration or by the event extraction process. More importantly, there is no indication of a time lapse or a gap in the story, so the standard forward-backward algorithm does not apply. To account for this, we allow the states to skip generating observations with some probability. This kind of HMMs, with insertions and gaps, have been considered previously in speech processing BIBREF7 and in computational biology BIBREF8 . We refine these models by allowing state-dependent missingness, without introducing additional “insert states” or “delete states” as in BIBREF8 . In this paper, we restrict our attention to the so-called “Left-to-Right HMMs” which have acyclic graphical structure with possible self-loops, as they support more efficient inference algorithms than general HMMs and suffice to model most of the natural scripts. We consider the problem of learning the structure and parameters of scripts in the form of HMMs from sequences of natural language sentences. Our solution to script learning is a novel bottom-up method for structure learning, called SEM-HMM, which is inspired by Bayesian Model Merging (BMM) BIBREF9 and Structural Expectation Maximization (SEM) BIBREF10 . It starts with a fully enumerated HMM representation of the event sequences and incrementally merges states and deletes edges to improve the posterior probability of the structure and the parameters given the data. We compare our approach to several informed baselines on many natural datasets and show its superior performance. We believe our work represents the first formalization of scripts that supports probabilistic inference, and paves the way for robust understanding of natural language texts.
In this paper we present the first formal representation of scripts as Hidden Markov Models (HMMs), which support robust inference and effective learning algorithms.
|
Do the Hidden Markov Models (HMMs) support robust inference and effective learning algorithms?
|
Yes, they are.
|
null | false
| null |
Classify each of the words as singular or plural: lions, bee, door, fence, ocean, lakes, people
|
Lions: Plural
Bee: Singular
Door: Singular
Fence: Singular
Ocean: Singular
Lakes: Plural
People: Plural
|
|
null | false
| null |
Is Dijkstras algorithm a greedy algorithm or dynamic programming algorithm
|
A greedy algorithm.
|
|
1703.05260
| false
| null |
In order to calculate inter-annotator agreement, a total of 30 stories from 6 scenarios were randomly chosen for parallel annotation by all 4 annotators after the first annotation phase. We checked the agreement on these data using Fleiss' Kappa BIBREF4 . The results are shown in Figure 4 and indicate moderate to substantial agreement BIBREF5 . Interestingly, if we calculated the Kappa only on the subset of cases that were annotated with script-specific event and participant labels by all annotators, results were better than those of the evaluation on all labeled instances (including also unrelated and related non-script events). This indicates one of the challenges of the annotation task: In many cases it is difficult to decide whether a particular event should be considered a central script event, or an event loosely related or unrelated to the script.
For coreference chain annotation, we calculated the percentage of pairs which were annotated by at least 3 annotators (qualified majority vote) compared to the set of those pairs annotated by at least one person (see Figure 4 ). We take the result of 90.5% between annotators to be a good agreement.
FLOAT SELECTED: Figure 4: Inter-annotator agreement statistics.
The results are shown in Figure 4 and indicate moderate to substantial agreement BIBREF5 .
We take the result of 90.5% between annotators to be a good agreement.
FLOAT SELECTED: Figure 4: Inter-annotator agreement statistics.
|
Did the annotators agreed and how much?
|
Moderate agreement of 0.64-0.68 Fleiss’ Kappa over event type labels, 0.77 Fleiss’ Kappa over participant labels, and good agreement of 90.5% over coreference information.
|
null | false
| null |
Which years was the Tour de France not raced?
|
The Tour was not contested in 1915, 1916, 1917 and 1918 due to World War 1. It was again postponed from 1940 through 1945 due to World War 2.
|
|
null | false
| null |
In the House is an American television sitcom that originally aired on NBC from April 10, 1995 and moved to UPN after its second season. The show starred LL Cool J as Marion Hill a former professional football player with the Oakland Raiders. Because of his financial predicament, he is forced to rent out most of the rooms in his house to newly divorced single mother Jackie Warren (Debbie Allen) and her two children, Tiffany (Maia Campbell) and Austin (Jeffery Wood). In the third season, Jackie and Austin move to Nashville, leaving Tiffany to stay with Marion in order to finish school in Los Angeles; Marion purchases a sports rehabilitation facility with the boisterous Tonia (Kim Wayans) and the pompous Maxwell (Alfonso Ribeiro), leading to the trio's attempts to work together despite their clashing personalities.
|
Given a reference text about In the House, when was it originally on TV and what channel played it?
|
In the House was aired on NBC on April 10, 1995
|
null | false
| null |
What books would you recommend someone read to improve their general knowledge of the world?
|
Here are some books that can help improve one's general knowledge of the world:
*Guns, Germs, and Steel: The Fates of Human Societies Revised by Jared Diamond
* The Blind Watchmaker: Why the Evidence of Evolution Reveals a Universe Without Design by Richard Dawkins
* The Righteous Mind: Why Good People Are Divided by Politics and Religion by Jon Haidt
* The Psychology of Money: Timeless lessons on wealth, greed, and happiness by Housel Morgan
* Complexity: The Emerging Science at the Edge of Order and Chaos by Mitch Waldrop
* Influence: The Psychology of Persuasion by Robert Cialdini
* Thinking, Fast and Slow by Daniel Kahneman
* How the Mind Works by Steven Pinker
* Darwin's Dangerous Idea: Evolution and the Meanings of Life by Daniel Dennett
* Against the Gods: The Remarkable Story of Risk by Peter L. Bernstein
* Moneyball by Michael Lewis
* The Halo Effect: ... and the Eight Other Business Delusions That Deceive by Phil Rosenzweig
|
|
null | false
| 161
|
The main goal of the present work is to predict the emotion of utterance within the dialogue. Following are four major difficulties we concern about:
The emotion of the utterances depends not only on the text but also on the interaction happened earlier.
The source of the two datasets are different. Friends is speech-based dialogues and EmotionPush is chat-based dialogues. It makes datasets possess different characteristics.
There are only $1,000$ dialogues in both training datasets which are not large enough for the stability of training a complex neural-based model.
The prediction targets (emotion labels) are highly unbalanced.
The proposed approach is summarized in Figure FIGREF3, which aims to overcome these challenges. The framework could be separated into three steps and described as follow:
The main goal of the present work is to predict the emotion of utterance within the dialogue. Following are four major difficulties we concern about: 1. The emotion of the utterances depends not only on the text but also on the interaction happened earlier. 2. The source of the two datasets are different. Friends is speech-based dialogues and EmotionPush is chat-based dialogues. It makes datasets possess different characteristics. 3. There are only 1, 000 dialogues in both training datasets which are not large enough for the stability of training a complex neural-based model. 4. The prediction targets (emotion labels) are highly unbalanced.
|
What are the difficulties in the current work?
|
1.The emotion of the utterances depends not only on the text but also on the interaction happened earlier. 2. The source of the two datasets are different. 3. There are only 1, 000 dialogues in both training datasets which are not large enough for the stability of training a complex neural-based model. 4. The prediction targets (emotion labels) are highly unbalanced.
|
null | false
| null |
What is the benefit of turmeric
|
Fights against Alzheimer's disease, and certain types of cancer, Acts as an anti-inflammatory and antibacterial, Enhances digestion and prevents the cold and flu.
|
|
null | false
| null |
Wang Labs was one of a large number of New England-based computer companies that faltered in the late 1980s and 1990s, marking the end of the Massachusetts Miracle. For instance, the struggling Digital Equipment Corporation also downsized in the 1990s and was acquired by Compaq.
A common view within the PC community is that Wang Labs failed because it specialized in computers designed specifically for word processing and did not foresee and could not compete with general-purpose personal computers with word-processing software in the 1980s. Word processing was not actually the mainstay of Wang's business by the time desktop computers began to gain in popularity. Although Wang manufactured desktops, its main business by the 1980s was its VS line of minicomputer and "midframe" systems. The market for these minicomputers was conquered by enhanced microcomputers like the Apple Macintosh and the Wintel PC and Sun, IBM, and Hewlett-Packard servers.
An Wang's insistence that his son, Fred Wang, succeed him contributed to the company's failure. Fred Wang was a business school graduate, "but by almost any definition", wrote Charles C. Kenney, "unsuited for the job in which his father had placed him." His assignment, first as head of research and development, then as president of the company, led to resignations by key R&D and business personnel. Amid declining revenues, John F. Cunningham, an 18-year employee of the firm, resigned as president and COO of Wang Labs to become chairman and chief executive of Computer Consoles Inc. Cunningham resigned due to disagreement with An Wang on how to pull the company out of the slump, as well as being upset that Fred Wang was positioned, nepotistically, as An Wang's successor.
One turning point occurred when Fred Wang was head of R&D. On October 4, 1983, Wang Laboratories announced fourteen major hardware and software products and promised dates of delivery. The announcement was well received, but even at the time, there were warning signs. According to Datamation, Wang announced "everything but the kitchen sink. And if you could attach the kitchen sink to a personal computer, they would announce that too." Very few of the products were close to completion, and many of them had not even been started. All were delivered late, if at all. In retrospect, this was referred to as the "vaporware announcement," and it hurt the credibility of Fred Wang and Wang Laboratories.
In 1986, Fred Wang, then 36 years old, was installed as president of Wang Laboratories. However, the company's fortunes continued to decline. Unlike most computer companies that funded their growth by issuing stock, An Wang had used debt to avoid further dilution of family control of the company. By August 1989, that debt was causing conflicts with its creditors. On August 4, 1989, An Wang fired his son. Richard W. Miller, who had been with the company since 1988, replaced him as the president of Wang Laboratories.
Miller announced in December 1989 that the company would start to embrace established software standards rather than use traditional proprietary designs. An Wang died in March 1990, and Miller took on the additional posts of chairman and CEO. The company underwent massive restructuring and eliminated its bank debt in August 1990, but it still ended the year with a record net loss.
In November 1990, Wang announced their first personal computers running Unix. In 1987, Wang developed a new typesetting system in conjunction with Arlington, MA-based Texet Corp. The system used Xerox printers and UNIX workstations from Sun, but the product vanished before coming to market, because few Wang employees could use or support UNIX. UNIX ran on the VS – Interactive Systems first ported IN/ix (their IBM 360 version of SYS5 UNIX) to run in a VSOS Virtual machine circa 1985, and then Wang engineers completed the port so that it ran "native" on the VS hardware soon thereafter – but performance was always sub-par as UNIX was never a good fit for the batch-mode nature of the VS hardware, and the line-at-a-time processing approach taken by the VS workstations; indeed, the workstation code had to be rewritten to bundle up each keystroke into a frame to be sent back to the host when running UNIX so that "tty" style processing could be implemented. PACE, which offered its data dictionary, excellent referential integrity, and speedy application development, was in the process of being ported to UNIX under the name OPEN Pace. A client-server RDBMS model built on the original product's ideology, OPEN Pace was demonstrated at the North American PACE User Group Conferences in both Boston and Chicago. OPEN Pace, along with a new Windows-based word processor called UpWord (which was at the time considered a strong contender to retake Wang's original market leadership from Microsoft), were touted as their new direction. However, after a marketing study[citation needed] suggested that it would require large capital investments in order to be viable competitors against Microsoft, both products were abandoned.
Ira Magaziner, who was brought in by Miller in 1990, proposed to take Wang out of the manufacture of computers altogether, and to go big into imaging software instead. In March 1991, the company introduced its Office 2000 marketing strategy, focusing on office productivity.
In June 1991, Wang started reselling IBM computers, in exchange for IBM investing in Wang stock. Wang hardware strategy to re-sell IBM RS/6000s also included further pursuit of UNIX software.
In August 1991, Wang won a suit against NEC and Toshiba claiming violation of Wang's patents on single in-line memory modules (SIMMs). The company still recorded a net loss for the 1991 fiscal year.
Wang Laboratories filed for bankruptcy protection on August 18, 1992, at a time when the company's attempted concession from proprietary to open systems was deemed by some analysts as "too little and too late
|
What is the story of the decline of Wang Computer?
|
Founder An Wang forced his experienced son Fred Wang was going to lead the company after his death.
|
null | false
| null |
Why are kids inquisitive?
|
Kids are very eager to understand the world around them and tend to ask lots of questions. There is a potential for these characteristics to reduce as kids grow older, either because of other social interactions or having a better understanding of their general environment.
|
|
null | false
| null |
What is the value of higher education?
|
Higher education offers opportunities for people to expand their knowledge and experience. Choosing to pursue education beyond mandatory schooling provides people the chance to receive in-depth training on topics that they enjoy, but also have more exposure to a breadth of various topics. While everyone does not need higher eduction in order to be successful, it does offer many ways for people to explore new interests, develop a mastery of skills, and even learn more about themselves. Higher education offers academic, as well as personal benefits. For example, attending a college or university offers the chance for people to receive mentorship from distinguished professors, meet new people from diverse backgrounds, and share their own ideas with others. There is great value in higher education for people interested in learning new skills before they start their career, advancing their skillset in their current profession, or venturing into a new vocation.
|
|
null | false
| 109
|
NLP research is gradually evolving from lexical to compositional semantics BIBREF10 through the adoption of novel meaning-preserving and context-aware paradigms such as convolutional networks BIBREF11 , recurrent belief networks BIBREF12 , statistical learning theory BIBREF13 , convolutional multiple kernel learning BIBREF14 , and commonsense reasoning BIBREF15 . But while other NLP tasks have been extensively investigated, sarcasm detection is a relatively new research topic which has gained increasing interest only recently, partly thanks to the rise of social media analytics and sentiment analysis. Sentiment analysis BIBREF16 and using multimodal information as a new trend BIBREF17 , BIBREF18 , BIBREF19 , BIBREF20 , BIBREF14 is a popular branch of NLP research that aims to understand sentiment of documents automatically using combination of various machine learning approaches BIBREF21 , BIBREF22 , BIBREF20 , BIBREF23 .
An early work in this field was done by BIBREF6 on a dataset of 6,600 manually annotated Amazon reviews using a kNN-classifier over punctuation-based and pattern-based features, i.e., ordered sequence of high frequency words. BIBREF1 used support vector machine (SVM) and logistic regression over a feature set of unigrams, dictionary-based lexical features and pragmatic features (e.g., emoticons) and compared the performance of the classifier with that of humans. BIBREF24 described a set of textual features for recognizing irony at a linguistic level, especially in short texts created via Twitter, and constructed a new model that was assessed along two dimensions: representativeness and relevance. BIBREF5 used the presence of a positive sentiment in close proximity of a negative situation phrase as a feature for sarcasm detection. BIBREF25 used the Balanced Window algorithm for classifying Dutch tweets as sarcastic vs. non-sarcastic; n-grams (uni, bi and tri) and intensifiers were used as features for classification.
BIBREF26 compared the performance of different classifiers on the Amazon review dataset using the imbalance between the sentiment expressed by the review and the user-given star rating. Features based on frequency (gap between rare and common words), written spoken gap (in terms of difference between usage), synonyms (based on the difference in frequency of synonyms) and ambiguity (number of words with many synonyms) were used by BIBREF3 for sarcasm detection in tweets. BIBREF9 proposed the use of implicit incongruity and explicit incongruity based features along with lexical and pragmatic features, such as emoticons and punctuation marks. Their method is very much similar to the method proposed by BIBREF5 except BIBREF9 used explicit incongruity features. Their method outperforms the approach by BIBREF5 on two datasets.
BIBREF8 compared the performance with different language-independent features and pre-processing techniques for classifying text as sarcastic and non-sarcastic. The comparison was done over three Twitter dataset in two different languages, two of these in English with a balanced and an imbalanced distribution and the third one in Czech. The feature set included n-grams, word-shape patterns, pointedness and punctuation-based features.
In this work, we use features extracted from a deep CNN for sarcasm detection. Some of the key differences between the proposed approach and existing methods include the use of a relatively smaller feature set, automatic feature extraction, the use of deep networks, and the adoption of pre-trained NLP models.
Some of the key differences between the proposed approach and existing methods include the use of a relatively smaller feature set, automatic feature extraction, the use of deep networks, and the adoption of pre-trained NLP models.
|
What are the key differences between the proposed approach and existing methods?
|
Include the use of a relatively smaller feature set, automatic feature extraction, the use of deep networks, and the adoption of pre-trained NLP models.
|
null | false
| null |
What is the time value of money?
|
The time value of money is a concept in finance that reflects that money in the future is usually worth a smaller amount than the same amount of money today. The reason for this is that money today could be invested and earn a return that would make it worth a larger amount in the future.
|
|
null | false
| null |
Flink provides a high-throughput, low-latency streaming engine as well as support for event-time processing and state management. Flink applications are fault-tolerant in the event of machine failure and support exactly-once semantics. Programs can be written in Java, Scala, Python, and SQL and are automatically compiled and optimized into dataflow programs that are executed in a cluster or cloud environment.
|
From the passage provided, extract the programming languages supported by Flink. Separate them with a comma.
|
Java, Scala, Python, SQL
|
null | false
| null |
What can you learn from living in a foreign country?
|
A person that lives outside their native country is called an expatriate. One of the challenges of living abroad may be learning a new language. It is also important to understand the culture, and be sensitive to local customs and traditions as these may differ significantly from those you are used to. You may experience new food and drink, and even when and where you eat may vary from your native country. It is possible that fashions and clothing may vary to accommodate a different climate or lifestyle. Overall, spending time living in a foreign country will broaden your general knowledge and encourage you to view things from an alternative perspective.
|
|
null | false
| null |
Zhejiang Geely Holding Group Co., Ltd (ZGH), commonly known as Geely (吉利 /ˈdʒiːliː/ "auspicious"), is a Chinese multinational automotive company headquartered in Hangzhou, Zhejiang. The company is privately held by Chinese billionaire entrepreneur Li Shufu. It was established in 1986 and entered the automotive industry in 1997 with its Geely Auto subsidiary. Geely Auto is currently the seventh largest automobile manufacturer in China, with 1.328 million sales in China in 2021. Globally, the group sold over 2.2 million cars in 2021, and over 17,926 plug-in electric vehicles in January 2022.
The company produces and sells vehicles under its own branding - such as Geely Auto, Geometry, Maple, and Zeekr - and under foreign-located subsidiaries - such as Volvo Cars, Polestar, Lynk & Co, Proton, and Lotus - as well as commercial only vehicles under the London EV Company, Ouling Auto and Farizon Auto brands. It also produces electric vehicles under some of the previously listed brandings, and motorcycles under its subsidiary Zhejiang Geely Ming Industrial (Jiming and Geely brands), Qianjiang Motorcycle (QJMotor and Keeway brands) and Benelli. In September 2022 the group acquired a 7.6% shareholding in Aston Martin Lagonda Global Holdings plc, the ultra-luxury British performance brand Aston Martin.
|
Which brands does Geely own?
|
Chinese automotive company Geely owns brands such as Geely Auto, Geometry, Maple, Zeekr, Volvo Cars, Polsetar, Lynk & Co, Proton, Lotus, London EV Company, Ouling Auto and Farizon Auto.
|
null | false
| null |
When Kennedy died in August 2009, he was the second-most senior member of the Senate (after President pro tempore Robert Byrd of West Virginia) and the third longest-serving senator of all time, behind Byrd and Strom Thurmond of South Carolina. Later that same year, he was passed by Daniel Inouye of Hawaii. Kennedy therefore held the record as the longest-serving Democratic member of Congress to solely serve as a senator until October 2021, when he was surpassed by fellow Democrat Patrick Leahy of Vermont.
|
Based on this passage about Ted Kennedy, at the time of his death, who were the three longest-serving senators of all time?
|
Based on this passage, Robert Byrd, Strom Thurmond and Kennedy were the three longest-serving senators of all time.
|
null | false
| null |
Classify each of the following as either a country or a continent: Africa, North America, Asia, South America, Asia, China, Argentina, Canada, Mexico, Japan, United States of America
|
Africa is a continent.
North America is a continent.
Asia is a continent.
South America is a continent.
Asia is a continent.
China is a country in Asia.
Argentina is a country in South America.
Canada is a country in North America.
Mexico is a country in South America.
Japan is a country in Asia.
United States of America is a country in North America.
|
|
null | false
| null |
Which one among the following three words is different to the others: Imaginative, astounding, incredible? Why?
|
Imaginative is different to astounding and incredible, as it implies something imagined or imagined; astounding and incredible both suggest something extraordinary.
|
|
null | false
| null |
Can I feed table food to dogs?
|
You cannot feed table food to dogs because dogs have different digestive systems. The human food is usually too rich and fatty which is difficult for dogs to digest and thus causing health issues for dogs. Some table food such as chocolate, raisins, grapes are even toxic to dogs.
|
|
null | false
| 0
|
Although Neural Machine Translation (NMT) has dominated recent research on translation tasks BIBREF0, BIBREF1, BIBREF2, NMT heavily relies on large-scale parallel data, resulting in poor performance on low-resource or zero-resource language pairs BIBREF3. Translation between these low-resource languages (e.g., Arabic$\rightarrow $Spanish) is usually accomplished with pivoting through a rich-resource language (such as English), i.e., Arabic (source) sentence is translated to English (pivot) first which is later translated to Spanish (target) BIBREF4, BIBREF5. However, the pivot-based method requires doubled decoding time and suffers from the propagation of translation errors.
One common alternative to avoid pivoting in NMT is transfer learning BIBREF6, BIBREF7, BIBREF8, BIBREF9 which leverages a high-resource pivot$\rightarrow $target model (parent) to initialize a low-resource source$\rightarrow $target model (child) that is further optimized with a small amount of available parallel data. Although this approach has achieved success in some low-resource language pairs, it still performs very poorly in extremely low-resource or zero-resource translation scenario. Specifically, BIBREF8 reports that without any child model training data, the performance of the parent model on the child test set is miserable.
In this work, we argue that the language space mismatch problem, also named domain shift problem BIBREF10, brings about the zero-shot translation failure in transfer learning. It is because transfer learning has no explicit training process to guarantee that the source and pivot languages share the same feature distributions, causing that the child model inherited from the parent model fails in such a situation. For instance, as illustrated in the left of Figure FIGREF1, the points of the sentence pair with the same semantics are not overlapping in source space, resulting in that the shared decoder will generate different translations denoted by different points in target space. Actually, transfer learning for NMT can be viewed as a multi-domain problem where each source language forms a new domain. Minimizing the discrepancy between the feature distributions of different source languages, i.e., different domains, will ensure the smooth transition between the parent and child models, as shown in the right of Figure FIGREF1. One way to achieve this goal is the fine-tuning technique, which forces the model to forget the specific knowledge from parent data and learn new features from child data. However, the domain shift problem still exists, and the demand of parallel child data for fine-tuning heavily hinders transfer learning for NMT towards the zero-resource setting.
In this paper, we explore the transfer learning in a common zero-shot scenario where there are a lot of source$\leftrightarrow $pivot and pivot$\leftrightarrow $target parallel data but no source$\leftrightarrow $target parallel data. In this scenario, we propose a simple but effective transfer approach, the key idea of which is to relieve the burden of the domain shift problem by means of cross-lingual pre-training. To this end, we firstly investigate the performance of two existing cross-lingual pre-training methods proposed by BIBREF11 in zero-shot translation scenario. Besides, a novel pre-training method called BRidge Language Modeling (BRLM) is designed to make full use of the source$\leftrightarrow $pivot bilingual data to obtain a universal encoder for different languages. Once the universal encoder is constructed, we only need to train the pivot$\rightarrow $target model and then test this model in source$\rightarrow $target direction directly. The main contributions of this paper are as follows:
We propose a new transfer learning approach for NMT which uses the cross-lingual language model pre-training to enable a high performance on zero-shot translation.
We propose a novel pre-training method called BRLM, which can effectively alleviates the distance between different source language spaces.
Our proposed approach significantly improves zero-shot translation performance, consistently surpassing pivoting and multilingual approaches. Meanwhile, the performance on supervised translation direction remains the same level or even better when using our method.
We propose a novel pre-training method called BRLM, which can effectively alleviates the distance between different source language spaces.
|
What novel pre-training method do the authors propose?
|
BRLM.
|
null | false
| 115
|
During the recording session, the participants read 739 sentences that were selected from the Wikipedia corpus provided by culotta2006integrating. This corpus was chosen because it provides annotations of semantic relations. We included seven of the originally defined relation types: political_affiliation, education, founder, wife/husband, job_title, nationality, and employer. The sentences were chosen in the same length range as ZuCo 1.0, and with similar Flesch reading ease scores. The dataset statistics are shown in Table TABREF2.
Of the 739 sentences, the participants read 349 sentences in a normal reading paradigm, and 390 sentences in a task-specific reading paradigm, in which they had to determine whether a certain relation type occurred in the sentence or not. Table TABREF3 shows the distribution of the different relation types in the sentences of the task-specific annotation paradigm.
Purposefully, there are 63 duplicates between the normal reading and the task-specific sentences (8% of all sentences). The intention of these duplicate sentences is to provide a set of sentences read twice by all participants with a different task in mind. Hence, this enables the comparison of eye-tracking and brain activity data when reading normally and when annotating specific relations (see examples in Section SECREF4).
Furthermore, there is also an overlap in the sentences between ZuCo 1.0 and ZuCo 2.0. 100 normal reading and 85 task-specific sentences recorded for this dataset were already recorded in ZuCo 1.0. This allows for comparisons between the different recording procedures (i.e. session-specific effects) and between more participants (subject-specific effects).
During the recording session, the participants read 739 sentences that were selected from the Wikipedia corpus provided by culotta2006integrating. This corpus was chosen because it provides annotations of semantic relations.
|
Why do they choose the Wikipedia corpus provided by culotta2006integrating in the recording session?
|
Because it provides annotations of semantic relations.
|
null | false
| 203
|
Neural language models BIBREF0 , BIBREF1 , BIBREF2 have become an essential component in several areas of natural language processing (NLP), such as machine translation, speech recognition and image captioning. They have also become a common benchmarking application in machine learning research on recurrent neural networks (RNN), because producing an accurate probabilistic model of human language is a very challenging task which requires all levels of linguistic analysis, from pragmatics to phonology, to be taken into account.
A typical language model is trained on text in a single language, and if one needs to model multiple languages the standard solution is to train a separate model for each language. This presupposes large quantities of monolingual data in each of the languages that needs to be covered and each model with its parameters is completely independent of any of the other models.
We propose instead to use a single model with real-valued vectors to indicate the language used, and to train this model with a large number of languages. We thus get a language model whose predictive distribution INLINEFORM0 is a continuous function of the language vector INLINEFORM1 , a property that is trivially extended to other neural NLP models. In this paper, we explore the “language space” containing these vectors, and in particular explore what happens when we move beyond the points representing the languages of the training corpus.
The motivation of combining languages into one single model is at least two-fold: First of all, languages are related and share many features and properties, a fact that is ignored when using independent models. The second motivation is data sparseness, an issue that heavily influences the reliability of data-driven models. Resources are scarce for most languages in the world (and also for most domains in otherwise well-supported languages), which makes it hard to train reasonable parameters. By combining data from many languages, we hope to mitigate this issue.
In contrast to related work, we focus on massively multilingual data sets to cover for the first time a substantial amount of the linguistic diversity in the world in a project related to data-driven language modeling. We do not presuppose any prior knowledge about language similarities and evolution and let the model discover relations on its own purely by looking at the data. The only supervision that is giving during training is a language identifier as a one-hot encoding. From that and the actual training examples, the system learns dense vector representations for each language included in our data set along with the character-level RNN parameters of the language model itself.
The motivation of combining languages into one single model is at least two-fold: First of all, languages are related and share many features and properties, a fact that is ignored when using independent models. The second motivation is data sparseness, an issue that heavily influences the reliability of data-driven models.
|
What is the motivation for combining languages into one single model?
|
Languages are related and share many features and properties, and data sparseness.
|
null | false
| null |
When was the United States officially recognized as an independent nation?
|
The United Kingdom officially recognized the US as an independent nation on September 3, 1783.
|
|
null | false
| null |
Clark won the AP Player of the Year, Naismith College Player of the Year, USBWA National Player of the Year and the John R. Wooden Award and the Wade Trophy. She was a unanimous first-team All-American for a second straight season, earning first-team recognition from the AP and USBWA and making the WBCA Coaches' All-America Team. Clark received the Nancy Lieberman award for her second straight season.
|
Extract the awards that Caitlin Clark won during her Junior year season at Iowa State.
|
AP Player of the Year, Naismith College Player of the year, USBWA National Player of the Year, John R. Wooden Award, Wade Trophy, Nancy Lieberman award
|
null | false
| null |
Tell me whether these are airplanes or boats : Cessna, Regal, Sunray, Learjet, Airbus, Jetski, Boeing, Cirrus, Mooney and Zodiac.
|
The following are airplanes : Cessna, LEarjet, Airbus, Boeing, Cirrus and Mooney.
The following are boats : Regal, Sunray and Zodiac.
|
|
null | false
| null |
Identify which instrument is string or percussion: Klong khaek, Sanshin
|
Sanshin is string, Klong khaek is percussion.
|
|
null | false
| 53
|
Medical search engines are an essential component for many online medical applications, such as online diagnosis systems and medical document databases. A typical online diagnosis system, for instance, relies on a medical search engine. The search engine takes as input a user query that describes some symptoms and then outputs clinical concept entries that provide relevant information to assist in diagnosing the problem. One challenge medical search engines face is the segmentation of individual clinical entities. When a user query consists of multiple clinical entities, a search engine would often fail to recognize them as separate entities. For example, the user query “fever joint pain weight loss headache” contains four separate clinical entities: “fever”, “joint pain”, “weight loss”, and “headache”. But when the search engine does not recognize them as separate entities and proceeds to retrieve results for each word in the query, it may find "pain" in body locations other than "joint pain", or it may miss "headache" altogether, for example. Some search engines allow the users to enter a single clinical concept by selecting from an auto-completion pick list. But this could also result in retrieving inaccurate or partial results and lead to poor user experience.
We want to improve the medical search engine so that it can accurately retrieve all the relevant clinical concepts mentioned in a user query, where relevant clinical concepts are defined with respect to the terminologies the search engine uses. The problem of extracting clinical concept mentions from a user query can be seen as a variant of the Concept Extraction (CE) task in the frequently-cited NLP challenges in healthcare, such as 2010 i2b2/VA BIBREF0 and 2013 ShARe/CLEF Task 1 BIBREF1. Both CE tasks in 2010 i2b2/VA and 2013 ShARe/CLEF Task 1 ask the participants to design an algorithm to tag a set of predefined entities of interest in clinical notes. These entity tagging tasks are also known as clinical Named Entity Recognition (NER). For example, the CE task in 2010 i2b2/VA defines three types of entities: “problem”, “treatment”, and “test”. The CE task in 2013 ShARe/CLEF defines various types of disorder such as “injury or poisoning”, "disease or syndrome”, etc. In addition to tagging, the CE task in 2013 ShARe/CLEF has an encoding component which requires selecting one and only one Concept Unique Identifier (CUI) from Systematized Nomenclature Of Medicine Clinical Terms (SNOMED-CT) for each disorder entity tagged. Our problem, similar to the CE task in 2013 ShARe/CLEF, also contains two sub-problems: tagging mentions of entities of interest (entity tagging), and selecting appropriate terms from a glossary to match the mentions (term matching). However, several major differences exist. First, compared to clinical notes, the user queries are much shorter, less technical, and often less coherent. Second, instead of encoding, we are dealing with term matching where we rank a few best terms that match an entity, instead of selecting only one. This is because the users who type the queries may not have a clear idea about what they are looking for, or could be laymen who know little terminology, it may be more helpful to provide a set of likely results and let the users choose. Third, the types of entities are different. Each medical search engine may have its own types of entities to tag. There is also one minor difference in the tagging scheme between our problem and the CE task in 2013 ShARe/CLEF - We limit our scope to dealing with entities of consecutive words and not disjoint entities . We use only Beginning, Inside, Outside (BIO) tags. Given the differences listed above, we need to customize a framework consisting of an entity tagging and term matching component for our CE problem.
The search engine takes as input a user query that describes some symptoms and then outputs clinical concept entries that provide relevant information to assist in diagnosing the problem.
|
How do medical search engines work?
|
The search engine takes as input a user query that describes some symptoms and then outputs clinical concept entries that provide relevant information to assist in diagnosing the problem.
|
null | false
| null |
Virgin Australia, the trading name of Virgin Australia Airlines Pty Ltd, is an Australian-based airline. It is the largest airline by fleet size to use the Virgin brand. It commenced services on 31 August 2000 as Virgin Blue, with two aircraft on a single route. It suddenly found itself as a major airline in Australia's domestic market after the collapse of Ansett Australia in September 2001. The airline has since grown to directly serve 32 cities in Australia, from hubs in Brisbane, Melbourne and Sydney.
|
When did Virgin Australia start operating?
|
Virgin Australia commenced services on 31 August 2000 as Virgin Blue, with two aircraft on a single route.
|
null | false
| null |
Tell me about Emily Procter
|
Emily Mallory Procter (born October 8, 1968) is an American actress and activist. She played Ainsley Hayes in the NBC political drama The West Wing (2000–2002; 2006) and Det. Calleigh Duquesne in the CBS police procedural drama CSI: Miami (2002–2012).
|
|
2001.08845
| true
| null |
In cooperation with the ruling regime, Weibo sets strict control over the content published under its service BIBREF0. According to Zhu et al. zhu-etal:2013, Weibo uses a variety of strategies to target censorable posts, ranging from keyword list filtering to individual user monitoring. Among all posts that are eventually censored, nearly 30% of them are censored within 5–30 minutes, and nearly 90% within 24 hours BIBREF1. We hypothesize that the former are done automatically, while the latter are removed by human censors.
In cooperation with the ruling regime, Weibo sets strict control over the content published under its service BIBREF0. According to Zhu et al. zhu-etal:2013, Weibo uses a variety of strategies to target censorable posts, ranging from keyword list filtering to individual user monitoring. Among all posts that are eventually censored, nearly 30% of them are censored within 5–30 minutes, and nearly 90% within 24 hours BIBREF1. We hypothesize that the former are done automatically, while the latter are removed by human censors.
|
Is is known whether Sina Weibo posts are censored by humans or some automatic classifier?
|
No.
|
null | false
| null |
Why are over partitioned tables poor data layout strategy
|
Over-partitioning creates physical boundaries beyond which the file compaction and z-ordering aren’t effective and you’re still left with the small files problems. With Delta, a better strategy is to use combination of partitioning and z-ordering to allow for better file layouts.
Would be helpful to understand the business requirements & design details behind these highly partitioned tables you have in parquet today
|
|
null | false
| null |
Anandapuram mandal is one of the 46 mandals in Visakhapatnam district of the Indian state of Andhra Pradesh. It is under the administration of Visakhapatnam revenue division and the headquarters are located at Anandapuram. Anandapuram mandal is bounded by Bheemunipatnam mandal to the east, Chinagadila to the south, Padmanabham to the north and Kothavalasa to the west.
|
What is Anandapuram mandal
|
Anandapuram mandal is one of the 46 mandals in Visakhapatnam district of the Indian state of Andhra Pradesh. It is under the administration of Visakhapatnam revenue division and the headquarters are located at Anandapuram.
|
null | false
| null |
What is the difference between and instructional designer and an elearning developer?
|
Instructional designers and elearning developers both play an important role in creating learning experiences. Instructional designers typically establish learning objectives in partnership with stakeholders, then write content based on the learning objectives. Usually, these learning objectives are skills-based and rely on an instructional designer's expertise to appropriately level the content and create activities based on task, skill and gap analysis. The content and activities created by an instructional designer is documented inside of a storyboard. The elearning developer then takes the storyboard and creates an interactive learning experience in software like Storyline 360 or Adobe Captivate. They carefully read the storyboard to understand where content belongs and how activities should be built. Elearning developers are usually more "junior" than instructional designers, because it takes less independent decision-making and expertise to be successful in their role.
|
|
null | false
| 133
|
The task of document quality assessment is to automatically assess a document according to some predefined inventory of quality labels. This can take many forms, including essay scoring (quality = language quality, coherence, and relevance to a topic), job application filtering (quality = suitability for role + visual/presentational quality of the application), or answer selection in community question answering (quality = actionability + relevance of the answer to the question). In the case of this paper, we focus on document quality assessment in two contexts: Wikipedia document quality classification, and whether a paper submitted to a conference was accepted or not.
Automatic quality assessment has obvious benefits in terms of time savings and tractability in contexts where the volume of documents is large. In the case of dynamic documents (possibly with multiple authors), such as in the case of Wikipedia, it is particularly pertinent, as any edit potentially has implications for the quality label of that document (and around 10 English Wikipedia documents are edited per second). Furthermore, when the quality assessment task is decentralized (as in the case of Wikipedia and academic paper assessment), quality criteria are often applied inconsistently by different people, where an automatic document quality assessment system could potentially reduce inconsistencies and enable immediate author feedback.
Current studies on document quality assessment mainly focus on textual features. For example, BIBREF0 examine features such as the article length and the number of headings to predict the quality class of a Wikipedia article. In contrast to these studies, in this paper, we propose to combine text features with visual features, based on a visual rendering of the document. Figure 1 illustrates our intuition, relative to Wikipedia articles. Without being able to read the text, we can tell that the article in Figure 1 has higher quality than Figure 1 , as it has a detailed infobox, extensive references, and a variety of images. Based on this intuition, we aim to answer the following question: can we achieve better accuracy on document quality assessment by complementing textual features with visual features?
Our visual model is based on fine-tuning an Inception V3 model BIBREF1 over visual renderings of documents, while our textual model is based on a hierarchical biLSTM. We further combine the two into a joint model. We perform experiments on two datasets: a Wikipedia dataset novel to this paper, and an arXiv dataset provided by BIBREF2 split into three sub-parts based on subject category. Experimental results on the visual renderings of documents show that implicit quality indicators, such as images and visual layout, can be captured by an image classifier, at a level comparable to a text classifier. When we combine the two models, we achieve state-of-the-art results over 3/4 of our datasets.
This paper makes the following contributions:
All code and data associated with this research will be released on publication.
In contrast to these studies, in this paper, we propose to combine text features with visual features, based on a visual rendering of the document.
|
In this paper, what features did the author propose to combine?
|
The purpose of this paper is to propose to combine text features with visual features, based on a visual rendering of the document.
|
null | false
| null |
What is the difference in the US between a stop sign and a traffic light?
|
A stop sign is a red, eight-sided metal sign that requires a driver to stop and then proceed when it is safe to do so. Many intersections have either two or four stop signs. When more than one car arrives at the intersection, the car that arrived first has the right of way. If more than one car arrives at the same time, then the car to the right has the right of way. An electric traffic light typically has three different signals: red indicates drivers should stop until the light changes; a yellow light indicates that the driver should stop if it is safe to do so because the signal is changing from red to green; and a green light means that drivers may proceed through the intersection without stopping.
|
|
null | false
| null |
what is fish oil?
|
Fish oil is produced from the tissues of fish. Fish that contain omega-3 fatty acids are the ones that are used to produce fish oil supplements. These fatty acids have been show to reduce inflammation in the body and may reduce your risk for heart disease, high blood pressure, and rheumatoid arthritis. Fish oil can be consumed daily as a morning or nightly supplement for support or enhance your immune health.
|
|
null | false
| null |
Give me the recipe for the traditional crepe
|
1. Put the floor, 300 grams, on an adequate plate and create a hole in the middle
2. Insert the hole 3 eggs, 3 tablespoons of sugar, 2 tablespoons of sunflower oil, and 50 grams of melted butter
3. Whisk softly and add gradually 60 centiliters of milk. The dough will become lightly dense
4. Add 5 centiliters of Rhum
5. Warm up a frying pan at low heat, add some sunflower oil, and a ladleful of crepe dough, and spread it out on the pan to wait until one side is cooked, then flip the crepe.
|
|
null | false
| null |
What is a CRM?
|
Customer relationship management is a process in which a business or other organization administers its interactions with customers, typically using data analysis to study large amounts of information.
|
|
null | false
| null |
Write a self evaluation with the goal of getting a promotion.
|
During my time at this company, I have contributed to a number of high priority projects. These projects have impacted the performance of the overall company in a positive way. I shown teamwork and collaborated cross-functionally with many coworkers to complete important workstreams. My performance in all aspects have been high and the feedback from peers have been positive. Given my contributions to the team and company, I believe I should be considered for a promotion.
|
|
null | false
| null |
Which episode of season two of Game of Thrones did Alik Sakharov direct?
|
Alik Sakharov directed the third episode of season 2 of Game of Thrones called "What is Dead May Never Die."
|
|
null | false
| null |
Describe a plan for a road trip across Northern Italy
|
Begin your trip in Turin, an important centre of Piedmont with a rich cultural history and host of the 2006 Winter Olympics. Next, move to Alba and visit the vineyards, taking day trips to towns such as Barolo and Barbaresco.
After this, you can continue moving east to Parma, a great place for the foodie searching for Prosciutto. Close by is Modena, home of the luxury automotive industry in Italy - Ferrari, Lamborghini, De Tomaso, Pagani and Maserati all have had (or still have) headquarters here.
Bologna is a fine place to finish your trip
|
|
null | false
| null |
Which is India's 3rd Company which has crossed $100 billion market capitalization in Intraday trading?
|
HDFC Bank Ltd
|
|
null | false
| null |
The group originated in 1987, when a concert choir program was having a variety night occurring at Fullerton College. When John Tebay, the choir director, suggested that Andre Peek (1st tenor and lead vocalist at that time) and Armando Fonseca (2nd tenor) perform together, both of them decided to take Mr. Tebays advice and form a quartet, an a cappella group. Two additional members along the way, Royce Reynolds (bass) and Todd Dixon (baritone)got together and the foursome gained notice. Since then The Alley Cats have performed for many audiences, ranging from elementary school fun nights to being Grand.prize.winners on the Gong Show. Performing on the Arsenio Hall show to The Tonight Show with Jay Leno, as well as opening for Leno at his Las Vegas shows. The Alley Cats's sound has been heard as well at The White House. The Alley Cats appeared in the Richard Simmons workout video Dance Your Pants Off! performed the song Celebration.
|
Give this paragraph about the Alley Cats a capella group, what famous talk show hosts have they performed for?
|
Jay Leno and Arsenio Hall
|
null | false
| 434
|
Stopping criterion: We stopped the firefly optimization when the latency of the best design found did not improve over the previous 1000 iterations, but we also made sure to run firefly optimization for at least 8000 iterations, to make sure that both the online and offline methods match in terms of the data budget. We also provide the convergence curves for firefly optimization on various single-application problems from Table 3 in Figure 9.
|
What is the stopping criteria for firefly?
|
We have now added this detail to Appendix B.1.1 (paragraph called “Stopping criterion”). In our experiments we ran the Firefly optimizer for atleast 8000 iterations and performed early stopping if the optimal solution found by the optimizer had not changed over the last 1000 iterations. This would typically mean stopping at 8000 iterations, since solutions found by the optimizer plateau after 6000 iterations or so (see convergence curves in Figure 8). We also ran online optimization for 4× as longer (i.e., 32000 iterations) for the t-RNN Enc, t-RNN Dec and MobilenetEdge applications in Table 18 and found that this made minimal difference on the solution found by the firefly.
|
null | false
| null |
what is love?
|
Love comes in many forms. Love can be between persons, love can be for an animal, or love can be for something that is fun or enjoyable like reading or writing. The feeling associated with love are often happiness, joy, fun and excitement. Love between persons can be done by showing affection, thoughtfully communicating, or doing a shared activity together. Love for an animal can be done by taking them for a walk or petting them. Love for something that is fun can be done by doing that activity over and over again. The opposite of love is hate and is associated more with feelings of despair.
|
|
null | false
| null |
The company was incorporated as Tesla Motors, Inc. on July 1, 2003, by Martin Eberhard and Marc Tarpenning. Eberhard and Tarpenning served as CEO and CFO, respectively. Eberhard said he wanted to build "a car manufacturer that is also a technology company", with its core technologies as "the battery, the computer software, and the proprietary motor".
Ian Wright was Tesla's third employee, joining a few months later. In February 2004, the company raised $7.5 million in series A funding, including $6.5 million from Elon Musk, who had received $100 million from the sale of his interest in PayPal two years earlier. Musk became the chairman of the board of directors and the largest shareholder of Tesla. J. B. Straubel joined Tesla in May 2004 as chief technical officer.
A lawsuit settlement agreed to by Eberhard and Tesla in September 2009 allows all five – Eberhard, Tarpenning, Wright, Musk, and Straubel – to call themselves co-founders.
|
Who was chairman of the board of directors of Tesla as of March 2004?
|
Elon Musk
|
2001.11381
| false
| null |
Corpus utilizados ::: Corpus 5KL
Este corpus fue constituido con aproximadamente 5 000 documentos (en su mayor parte libros) en español. Los documentos originales, en formatos heterogéneos, fueron procesados para crear un único documento codificado en utf8. Las frases fueron segmentadas automáticamente, usando un programa en PERL 5.0 y expresiones regulares, para obtener una frase por línea.
Las características del corpus 5KL se encuentran en la Tabla TABREF4. Este corpus es empleado para el entrenamiento de los modelos de aprendizaje profundo (Deep Learning, Sección SECREF4).
El corpus literario 5KL posee la ventaja de ser muy extenso y adecuado para el aprendizaje automático. Tiene sin embargo, la desventaja de que no todas las frases son necesariamente “frases literarias”. Muchas de ellas son frases de lengua general: estas frases a menudo otorgan una fluidez a la lectura y proporcionan los enlaces necesarios a las ideas expresadas en las frases literarias.
Otra desventaja de este corpus es el ruido que contiene. El proceso de segmentación puede producir errores en la detección de fronteras de frases. También los números de página, capítulos, secciones o índices producen errores. No se realizó ningún proceso manual de verificación, por lo que a veces se introducen informaciones indeseables: copyrights, datos de la edición u otros. Estas son, sin embargo, las condiciones que presenta un corpus literario real.
Corpus utilizados ::: Corpus 8KF
Un corpus heterogéneo de casi 8 000 frases literarias fue constituido manualmente a partir de poemas, discursos, citas, cuentos y otras obras. Se evitaron cuidadosamente las frases de lengua general, y también aquellas demasiado cortas ($N \le 3$ palabras) o demasiado largas ($N \ge 30$ palabras). El vocabulario empleado es complejo y estético, además que el uso de ciertas figuras literarias como la rima, la anáfora, la metáfora y otras pueden ser observadas en estas frases.
Las características del corpus 8KF se muestran en la Tabla TABREF6. Este corpus fue utilizado principalmente en los dos modelos generativos: modelo basado en cadenas de Markov (Sección SECREF13) y modelo basado en la generación de Texto enlatado (Canned Text, Sección SECREF15).
Corpus utilizados ::: Corpus 5KL
Este corpus fue constituido con aproximadamente 5 000 documentos (en su mayor parte libros) en español. Los documentos originales, en formatos heterogéneos, fueron procesados para crear un único documento codificado en utf8. Las frases fueron segmentadas automáticamente, usando un programa en PERL 5.0 y expresiones regulares, para obtener una frase por línea.
Las características del corpus 5KL se encuentran en la Tabla TABREF4. Este corpus es empleado para el entrenamiento de los modelos de aprendizaje profundo (Deep Learning, Sección SECREF4).
El corpus literario 5KL posee la ventaja de ser muy extenso y adecuado para el aprendizaje automático. Tiene sin embargo, la desventaja de que no todas las frases son necesariamente “frases literarias”. Muchas de ellas son frases de lengua general: estas frases a menudo otorgan una fluidez a la lectura y proporcionan los enlaces necesarios a las ideas expresadas en las frases literarias.
Otra desventaja de este corpus es el ruido que contiene. El proceso de segmentación puede producir errores en la detección de fronteras de frases. También los números de página, capítulos, secciones o índices producen errores. No se realizó ningún proceso manual de verificación, por lo que a veces se introducen informaciones indeseables: copyrights, datos de la edición u otros. Estas son, sin embargo, las condiciones que presenta un corpus literario real.
Corpus utilizados ::: Corpus 8KF
Un corpus heterogéneo de casi 8 000 frases literarias fue constituido manualmente a partir de poemas, discursos, citas, cuentos y otras obras. Se evitaron cuidadosamente las frases de lengua general, y también aquellas demasiado cortas ($N \le 3$ palabras) o demasiado largas ($N \ge 30$ palabras). El vocabulario empleado es complejo y estético, además que el uso de ciertas figuras literarias como la rima, la anáfora, la metáfora y otras pueden ser observadas en estas frases.
Las características del corpus 8KF se muestran en la Tabla TABREF6. Este corpus fue utilizado principalmente en los dos modelos generativos: modelo basado en cadenas de Markov (Sección SECREF13) y modelo basado en la generación de Texto enlatado (Canned Text, Sección SECREF15).
|
What datasets are used?
|
The answers are shown as follows:
* Corpus 5KL
* Corpus 8KF
|
null | false
| null |
On what date was the country of Belize granted independence?
|
Belize was granted independence from the British Empire on September 21, 1981.
|
|
null | false
| null |
What is Methode Traditionalle?
|
Methode Traditionalle is a method of producing sparkling wine. It was first created in 1531 and involves double fermentation of wine. The second fermentation creates carbon dioxide which creates the bubbles in the sparkling wine.
|
|
null | false
| 121
|
As outlined in Section SECREF2, most previous work BIBREF28, BIBREF12, BIBREF13, BIBREF10, BIBREF29 make use of flat encoders that do not exploit the data structure. To keep the semantics of each element from the data-structure, we propose a hierarchical encoder which relies on two modules. The first one (module A in Figure FIGREF11) is called low-level encoder and encodes entities on the basis of their records; the second one (module B), called high-level encoder, encodes the data-structure on the basis of its underlying entities. In the low-level encoder, the traditional embedding layer is replaced by a record embedding layer as in BIBREF11, BIBREF12, BIBREF10. We present in what follows the record embedding layer and introduce our two hierarchical modules.
The first layer of the network consists in learning two embedding matrices to embed the record keys and values. Keys $k_{i,j}$ are embedded to $\mathbf {k}_{i,j} \in \mathbb {R}^{d}$ and values $v_{i,j}$ to $\mathbf {v}_{i,j} \in \mathbb {R}^{d}$, with $d$ the size of the embedding. As in previous work BIBREF11, BIBREF12, BIBREF10, each record embedding $\mathbf {r}_{i,j}$ is computed by a linear projection on the concatenation $[\mathbf {k}_{i,j}$; $\mathbf {v}_{i,j}]$ followed by a non linearity:
where $\mathbf {W}_r \in \mathbb {R}^{2d \times d}$ and $\mathbf {b}_r \in \mathbb {R}^{d}$ are learnt parameters.
The low-level encoder aims at encoding a collection of records belonging to the same entity while the high-level encoder encodes the whole set of entities. Both the low-level and high-level encoders consider their input elements as unordered. We use the Transformer architecture from BIBREF21. For each encoder, we have the following peculiarities:
the Low-level encoder encodes each entity $e_i$ on the basis of its record embeddings $\mathbf {r}_{i,j}$. Each record embedding $\mathbf {r}_{i,j}$ is compared to other record embeddings to learn its final hidden representation $\mathbf {h}_{i,j}$. Furthermore, we add a special record [ENT] for each entity, illustrated in Figure FIGREF11 as the last record. Since entities might have a variable number of records, this token allows to aggregate final hidden record representations $\lbrace \mathbf {h}_{i,j}\rbrace _{j=1}^{J_i}$ in a fixed-sized representation vector $\mathbf {h}_{i}$.
the High-level encoder encodes the data-structure on the basis of its entity representation $\mathbf {h}_{i}$. Similarly to the Low-level encoder, the final hidden state $\mathbf {e_i}$ of an entity is computed by comparing entity representation $\mathbf {h}_{i}$ with each others. The data-structure representation $\mathbf {z}$ is computed as the mean of these entity representations, and is used for the decoder initialization.
Both the low-level and high-level encoders consider their input elements as unordered.
|
In Record Embedding Layer, is the input elements of the high-level encoder ordered?
|
No.
|
null | false
| null |
What is the best way to clean a cast iron pan that is heavily soiled?
|
Use a scrub brush or scrub sponge, but not one with any kind of metal in it. Contrary to popular opinion, you should be able to use soap if your pan is properly seasoned, because proper seasoning results from polymerization of oils across and to the pan itself, rather than just oil sitting on the surface of the pan. If necessary, you can re-season the pan, but this is usually excessive.
|
|
null | false
| null |
Solveig Margrete Fiske (born 26 October 1952) is a theologian and bishop emerita in the Church of Norway, serving as the bishop of Hamar between 2006 and 2022, when she retired.
Originally from Frei in Møre og Romsdal, she graduated from the Free Faculty of Theology (now the MF Norwegian School of Theology) in 1980, and completed her practical examinations there in 1981. She briefly worked as a consultant for the Church Council in 1981, followed by an assignment to the parish in Elverum as a chaplain between 1982 and 1994. In 1994, she became the parish priest in Løten, a position she held for 12 years.
|
Who is Solveig Margrete Fiske
|
Solveig Margrete Fiske (born 26 October 1952) is a theologian and bishop emerita in the Church of Norway, serving as the bishop of Hamar between 2006 and 2022, when she retired.
|
null | false
| null |
What are some things that I could gift a curious toddler?
|
Toddlers are at an age where they are very curious and are learning by observing and playing with things around them. Some toys that you could gift a curious toddler are:
* Building blocks
* Picture books
* Simple puzzles
* A scooter
* Toys that play nursery rhymes
* Balls of different sizes
* Bubble gun
... and many more.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.