Datasets:
DataHammer
/
SciEvents

Dataset card Data Studio Files
xet
 Community
1
venue
stringclasses
1 value
title
stringlengths
18
162
abstract
stringlengths
252
1.89k
doc_id
stringlengths
32
32
publication_year
int64
2.02k
2.02k
sentences
listlengths
1
13
events
listlengths
1
24
document
listlengths
50
348
ACL
AND does not mean OR: Using Formal Languages to Study Language Models’ Representations
A current open question in natural language processing is to what extent language models, which are trained with access only to the form of language, are able to capture the meaning of language. This question is challenging to answer in general, as there is no clear line between meaning and form, but rather meaning con...
6afda9deb3f6c7dfc43c6a12d4597c5e
2,021
[ "a current open question in natural language processing is to what extent language models , which are trained with access only to the form of language , are able to capture the meaning of language .", "this question is challenging to answer in general , as there is no clear line between meaning and form , but rat...
[ { "event_type": "ITT", "arguments": [ { "text": "language models", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "language", "models" ], "offsets": [ 12, 13 ] } ], "trigger": { ...
[ "a", "current", "open", "question", "in", "natural", "language", "processing", "is", "to", "what", "extent", "language", "models", ",", "which", "are", "trained", "with", "access", "only", "to", "the", "form", "of", "language", ",", "are", "able", "to", "c...
ACL
Bilingual Lexicon Induction via Unsupervised Bitext Construction and Word Alignment
Bilingual lexicons map words in one language to their translations in another, and are typically induced by learning linear projections to align monolingual word embedding spaces. In this paper, we show it is possible to produce much higher quality lexicons with methods that combine (1) unsupervised bitext mining and (...
b475197739d1f5750b99ad2202e9626a
2,021
[ "bilingual lexicons map words in one language to their translations in another , and are typically induced by learning linear projections to align monolingual word embedding spaces .", "in this paper , we show it is possible to produce much higher quality lexicons with methods that combine ( 1 ) unsupervised bite...
[ { "event_type": "ITT", "arguments": [ { "text": "bilingual lexicons", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "bilingual", "lexicons" ], "offsets": [ 0, 1 ] } ], "trigger"...
[ "bilingual", "lexicons", "map", "words", "in", "one", "language", "to", "their", "translations", "in", "another", ",", "and", "are", "typically", "induced", "by", "learning", "linear", "projections", "to", "align", "monolingual", "word", "embedding", "spaces", "...
ACL
Knowledge-Enriched Event Causality Identification via Latent Structure Induction Networks
Identifying causal relations of events is an important task in natural language processing area. However, the task is very challenging, because event causality is usually expressed in diverse forms that often lack explicit causal clues. Existing methods cannot handle well the problem, especially in the condition of lac...
9028fc83291fb95cac1f1fd4c843fc89
2,021
[ "identifying causal relations of events is an important task in natural language processing area .", "however , the task is very challenging , because event causality is usually expressed in diverse forms that often lack explicit causal clues .", "existing methods cannot handle well the problem , especially in ...
[ { "event_type": "ITT", "arguments": [ { "text": "natural language processing", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "natural", "language", "processing" ], "offsets": [ 10, 11, ...
[ "identifying", "causal", "relations", "of", "events", "is", "an", "important", "task", "in", "natural", "language", "processing", "area", ".", "however", ",", "the", "task", "is", "very", "challenging", ",", "because", "event", "causality", "is", "usually", "e...
ACL
Towards Unsupervised Language Understanding and Generation by Joint Dual Learning
In modular dialogue systems, natural language understanding (NLU) and natural language generation (NLG) are two critical components, where NLU extracts the semantics from the given texts and NLG is to construct corresponding natural language sentences based on the input semantic representations. However, the dual prope...
e3c2fb4e0d721e00e172a8b0f6847fa2
2,020
[ "in modular dialogue systems , natural language understanding ( nlu ) and natural language generation ( nlg ) are two critical components , where nlu extracts the semantics from the given texts and nlg is to construct corresponding natural language sentences based on the input semantic representations .", "howeve...
[ { "event_type": "ITT", "arguments": [ { "text": "natural language understanding", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "natural", "language", "understanding" ], "offsets": [ 5, 6, ...
[ "in", "modular", "dialogue", "systems", ",", "natural", "language", "understanding", "(", "nlu", ")", "and", "natural", "language", "generation", "(", "nlg", ")", "are", "two", "critical", "components", ",", "where", "nlu", "extracts", "the", "semantics", "fro...
ACL
Verb Metaphor Detection via Contextual Relation Learning
Correct natural language understanding requires computers to distinguish the literal and metaphorical senses of a word. Recent neu- ral models achieve progress on verb metaphor detection by viewing it as sequence labeling. In this paper, we argue that it is appropriate to view this task as relation classification betwe...
2a24568e53966df413355bc2a1d402f6
2,021
[ "correct natural language understanding requires computers to distinguish the literal and metaphorical senses of a word .", "recent neu - ral models achieve progress on verb metaphor detection by viewing it as sequence labeling .", "in this paper , we argue that it is appropriate to view this task as relation c...
[ { "event_type": "ITT", "arguments": [ { "text": "natural language understanding", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "natural", "language", "understanding" ], "offsets": [ 1, 2, ...
[ "correct", "natural", "language", "understanding", "requires", "computers", "to", "distinguish", "the", "literal", "and", "metaphorical", "senses", "of", "a", "word", ".", "recent", "neu", "-", "ral", "models", "achieve", "progress", "on", "verb", "metaphor", "d...
ACL
LSTMEmbed: Learning Word and Sense Representations from a Large Semantically Annotated Corpus with Long Short-Term Memories
While word embeddings are now a de facto standard representation of words in most NLP tasks, recently the attention has been shifting towards vector representations which capture the different meanings, i.e., senses, of words. In this paper we explore the capabilities of a bidirectional LSTM model to learn representati...
22703694469a0433df8472a4435f3f26
2,019
[ "while word embeddings are now a de facto standard representation of words in most nlp tasks , recently the attention has been shifting towards vector representations which capture the different meanings , i . e . , senses , of words .", "in this paper we explore the capabilities of a bidirectional lstm model to ...
[ { "event_type": "ITT", "arguments": [ { "text": "vector representations", "nugget_type": "FEA", "argument_type": "Target", "tokens": [ "vector", "representations" ], "offsets": [ 24, 25 ] } ], ...
[ "while", "word", "embeddings", "are", "now", "a", "de", "facto", "standard", "representation", "of", "words", "in", "most", "nlp", "tasks", ",", "recently", "the", "attention", "has", "been", "shifting", "towards", "vector", "representations", "which", "capture"...
ACL
ConvoSumm: Conversation Summarization Benchmark and Improved Abstractive Summarization with Argument Mining
While online conversations can cover a vast amount of information in many different formats, abstractive text summarization has primarily focused on modeling solely news articles. This research gap is due, in part, to the lack of standardized datasets for summarizing online discussions. To address this gap, we design a...
090cf3409b965753a4be0d9f79ee115d
2,021
[ "while online conversations can cover a vast amount of information in many different formats , abstractive text summarization has primarily focused on modeling solely news articles .", "this research gap is due , in part , to the lack of standardized datasets for summarizing online discussions .", "to address t...
[ { "event_type": "ITT", "arguments": [ { "text": "abstractive text summarization", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "abstractive", "text", "summarization" ], "offsets": [ 15, 16...
[ "while", "online", "conversations", "can", "cover", "a", "vast", "amount", "of", "information", "in", "many", "different", "formats", ",", "abstractive", "text", "summarization", "has", "primarily", "focused", "on", "modeling", "solely", "news", "articles", ".", ...
ACL
Weight Poisoning Attacks on Pretrained Models
Recently, NLP has seen a surge in the usage of large pre-trained models. Users download weights of models pre-trained on large datasets, then fine-tune the weights on a task of their choice. This raises the question of whether downloading untrusted pre-trained weights can pose a security threat. In this paper, we show ...
9476579fea44c819f0e002fc06460da1
2,020
[ "recently , nlp has seen a surge in the usage of large pre - trained models .", "users download weights of models pre - trained on large datasets , then fine - tune the weights on a task of their choice .", "this raises the question of whether downloading untrusted pre - trained weights can pose a security thre...
[ { "event_type": "ITT", "arguments": [ { "text": "large pre - trained models", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "large", "pre", "-", "trained", "models" ], "offsets": [ ...
[ "recently", ",", "nlp", "has", "seen", "a", "surge", "in", "the", "usage", "of", "large", "pre", "-", "trained", "models", ".", "users", "download", "weights", "of", "models", "pre", "-", "trained", "on", "large", "datasets", ",", "then", "fine", "-", ...
ACL
Learning to Generalize to More: Continuous Semantic Augmentation for Neural Machine Translation
The principal task in supervised neural machine translation (NMT) is to learn to generate target sentences conditioned on the source inputs from a set of parallel sentence pairs, and thus produce a model capable of generalizing to unseen instances. However, it is commonly observed that the generalization performance of...
28d66b18fbdb325636a374e87a728db8
2,022
[ "the principal task in supervised neural machine translation ( nmt ) is to learn to generate target sentences conditioned on the source inputs from a set of parallel sentence pairs , and thus produce a model capable of generalizing to unseen instances .", "however , it is commonly observed that the generalization...
[ { "event_type": "ITT", "arguments": [ { "text": "neural machine translation", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "neural", "machine", "translation" ], "offsets": [ 5, 6, ...
[ "the", "principal", "task", "in", "supervised", "neural", "machine", "translation", "(", "nmt", ")", "is", "to", "learn", "to", "generate", "target", "sentences", "conditioned", "on", "the", "source", "inputs", "from", "a", "set", "of", "parallel", "sentence",...
ACL
Give Me More Feedback II: Annotating Thesis Strength and Related Attributes in Student Essays
While the vast majority of existing work on automated essay scoring has focused on holistic scoring, researchers have recently begun work on scoring specific dimensions of essay quality. Nevertheless, progress on dimension-specific essay scoring is limited in part by the lack of annotated corpora. To facilitate advance...
e23e28022e3b8490a571dd794c4700df
2,019
[ "while the vast majority of existing work on automated essay scoring has focused on holistic scoring , researchers have recently begun work on scoring specific dimensions of essay quality .", "nevertheless , progress on dimension - specific essay scoring is limited in part by the lack of annotated corpora .", "...
[ { "event_type": "ITT", "arguments": [ { "text": "automated essay scoring", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "automated", "essay", "scoring" ], "offsets": [ 8, 9, 10 ...
[ "while", "the", "vast", "majority", "of", "existing", "work", "on", "automated", "essay", "scoring", "has", "focused", "on", "holistic", "scoring", ",", "researchers", "have", "recently", "begun", "work", "on", "scoring", "specific", "dimensions", "of", "essay",...
ACL
Transformer-XL: Attentive Language Models beyond a Fixed-Length Context
Transformers have a potential of learning longer-term dependency, but are limited by a fixed-length context in the setting of language modeling. We propose a novel neural architecture Transformer-XL that enables learning dependency beyond a fixed length without disrupting temporal coherence. It consists of a segment-le...
06007a191b140185a0fac49b7ad67005
2,019
[ "transformers have a potential of learning longer - term dependency , but are limited by a fixed - length context in the setting of language modeling .", "we propose a novel neural architecture transformer - xl that enables learning dependency beyond a fixed length without disrupting temporal coherence .", "it ...
[ { "event_type": "RWF", "arguments": [ { "text": "transformers", "nugget_type": "MOD", "argument_type": "Concern", "tokens": [ "transformers" ], "offsets": [ 0 ] }, { "text": "limited", "nugget_type"...
[ "transformers", "have", "a", "potential", "of", "learning", "longer", "-", "term", "dependency", ",", "but", "are", "limited", "by", "a", "fixed", "-", "length", "context", "in", "the", "setting", "of", "language", "modeling", ".", "we", "propose", "a", "n...
ACL
Beyond Laurel/Yanny: An Autoencoder-Enabled Search for Polyperceivable Audio
The famous “laurel/yanny” phenomenon references an audio clip that elicits dramatically different responses from different listeners. For the original clip, roughly half the population hears the word “laurel,” while the other half hears “yanny.” How common are such “polyperceivable” audio clips? In this paper we apply ...
c0ef5336b09e1ccd54912c9f26862a52
2,021
[ "the famous “ laurel / yanny ” phenomenon references an audio clip that elicits dramatically different responses from different listeners .", "for the original clip , roughly half the population hears the word “ laurel , ” while the other half hears “ yanny . ” how common are such “ polyperceivable ” audio clips ...
[ { "event_type": "ITT", "arguments": [ { "text": "“ polyperceivable ” audio clips", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "“", "polyperceivable", "”", "audio", "clips" ], "offsets": ...
[ "the", "famous", "“", "laurel", "/", "yanny", "”", "phenomenon", "references", "an", "audio", "clip", "that", "elicits", "dramatically", "different", "responses", "from", "different", "listeners", ".", "for", "the", "original", "clip", ",", "roughly", "half", ...
ACL
Know More about Each Other: Evolving Dialogue Strategy via Compound Assessment
In this paper, a novel Generation-Evaluation framework is developed for multi-turn conversations with the objective of letting both participants know more about each other. For the sake of rational knowledge utilization and coherent conversation flow, a dialogue strategy which controls knowledge selection is instantiat...
69c1c9b44dec7f80fb0ec3ecf6899047
2,019
[ "in this paper , a novel generation - evaluation framework is developed for multi - turn conversations with the objective of letting both participants know more about each other .", "for the sake of rational knowledge utilization and coherent conversation flow , a dialogue strategy which controls knowledge select...
[ { "event_type": "PRP", "arguments": [ { "text": "generation - evaluation framework", "nugget_type": "APP", "argument_type": "Content", "tokens": [ "generation", "-", "evaluation", "framework" ], "offsets": [ ...
[ "in", "this", "paper", ",", "a", "novel", "generation", "-", "evaluation", "framework", "is", "developed", "for", "multi", "-", "turn", "conversations", "with", "the", "objective", "of", "letting", "both", "participants", "know", "more", "about", "each", "othe...
ACL
Named Entity Recognition without Labelled Data: A Weak Supervision Approach
Named Entity Recognition (NER) performance often degrades rapidly when applied to target domains that differ from the texts observed during training. When in-domain labelled data is available, transfer learning techniques can be used to adapt existing NER models to the target domain. But what should one do when there i...
50cc365e1b70694fa691bf7ae26c7cef
2,020
[ "named entity recognition ( ner ) performance often degrades rapidly when applied to target domains that differ from the texts observed during training .", "when in - domain labelled data is available , transfer learning techniques can be used to adapt existing ner models to the target domain .", "but what shou...
[ { "event_type": "RWF", "arguments": [ { "text": "degrades", "nugget_type": "WEA", "argument_type": "Fault", "tokens": [ "degrades" ], "offsets": [ 8 ] }, { "text": "performance", "nugget_type": "TAK...
[ "named", "entity", "recognition", "(", "ner", ")", "performance", "often", "degrades", "rapidly", "when", "applied", "to", "target", "domains", "that", "differ", "from", "the", "texts", "observed", "during", "training", ".", "when", "in", "-", "domain", "label...
ACL
A Monolingual Approach to Contextualized Word Embeddings for Mid-Resource Languages
We use the multilingual OSCAR corpus, extracted from Common Crawl via language classification, filtering and cleaning, to train monolingual contextualized word embeddings (ELMo) for five mid-resource languages. We then compare the performance of OSCAR-based and Wikipedia-based ELMo embeddings for these languages on the...
9871351bc5f4c1aebb0a02b5f38ec839
2,020
[ "we use the multilingual oscar corpus , extracted from common crawl via language classification , filtering and cleaning , to train monolingual contextualized word embeddings ( elmo ) for five mid - resource languages .", "we then compare the performance of oscar - based and wikipedia - based elmo embeddings for ...
[ { "event_type": "WKS", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Researcher", "tokens": [ "we" ], "offsets": [ 0 ] }, { "text": "train", "nugget_type": "E-PUR", "...
[ "we", "use", "the", "multilingual", "oscar", "corpus", ",", "extracted", "from", "common", "crawl", "via", "language", "classification", ",", "filtering", "and", "cleaning", ",", "to", "train", "monolingual", "contextualized", "word", "embeddings", "(", "elmo", ...
ACL
Learning Span-Level Interactions for Aspect Sentiment Triplet Extraction
Aspect Sentiment Triplet Extraction (ASTE) is the most recent subtask of ABSA which outputs triplets of an aspect target, its associated sentiment, and the corresponding opinion term. Recent models perform the triplet extraction in an end-to-end manner but heavily rely on the interactions between each target word and o...
ef713c71fb25b074f849f97ba7d2c188
2,021
[ "aspect sentiment triplet extraction ( aste ) is the most recent subtask of absa which outputs triplets of an aspect target , its associated sentiment , and the corresponding opinion term .", "recent models perform the triplet extraction in an end - to - end manner but heavily rely on the interactions between eac...
[ { "event_type": "ITT", "arguments": [ { "text": "aspect sentiment triplet extraction", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "aspect", "sentiment", "triplet", "extraction" ], "offsets": [ ...
[ "aspect", "sentiment", "triplet", "extraction", "(", "aste", ")", "is", "the", "most", "recent", "subtask", "of", "absa", "which", "outputs", "triplets", "of", "an", "aspect", "target", ",", "its", "associated", "sentiment", ",", "and", "the", "corresponding",...
ACL
Can vectors read minds better than experts? Comparing data augmentation strategies for the automated scoring of children’s mindreading ability
In this paper we implement and compare 7 different data augmentation strategies for the task of automatic scoring of children’s ability to understand others’ thoughts, feelings, and desires (or “mindreading”). We recruit in-domain experts to re-annotate augmented samples and determine to what extent each strategy prese...
e7fda94261ca73f71111d89226063e20
2,021
[ "in this paper we implement and compare 7 different data augmentation strategies for the task of automatic scoring of children ’ s ability to understand others ’ thoughts , feelings , and desires ( or “ mindreading ” ) .", "we recruit in - domain experts to re - annotate augmented samples and determine to what ex...
[ { "event_type": "WKS", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Researcher", "tokens": [ "we" ], "offsets": [ 3 ] }, { "text": "7 different data augmentation strategies", ...
[ "in", "this", "paper", "we", "implement", "and", "compare", "7", "different", "data", "augmentation", "strategies", "for", "the", "task", "of", "automatic", "scoring", "of", "children", "’", "s", "ability", "to", "understand", "others", "’", "thoughts", ",", ...
ACL
Dynamic Memory Induction Networks for Few-Shot Text Classification
This paper proposes Dynamic Memory Induction Networks (DMIN) for few-short text classification. The model develops a dynamic routing mechanism over static memory, enabling it to better adapt to unseen classes, a critical capability for few-short classification. The model also expands the induction process with supervis...
53374dd4e5065382ca44e2cfd15ed51f
2,020
[ "this paper proposes dynamic memory induction networks ( dmin ) for few - short text classification .", "the model develops a dynamic routing mechanism over static memory , enabling it to better adapt to unseen classes , a critical capability for few - short classification .", "the model also expands the induct...
[ { "event_type": "PRP", "arguments": [ { "text": "dynamic memory induction networks", "nugget_type": "APP", "argument_type": "Content", "tokens": [ "dynamic", "memory", "induction", "networks" ], "offsets": [ ...
[ "this", "paper", "proposes", "dynamic", "memory", "induction", "networks", "(", "dmin", ")", "for", "few", "-", "short", "text", "classification", ".", "the", "model", "develops", "a", "dynamic", "routing", "mechanism", "over", "static", "memory", ",", "enabli...
ACL
A Multi-Perspective Architecture for Semantic Code Search
The ability to match pieces of code to their corresponding natural language descriptions and vice versa is fundamental for natural language search interfaces to software repositories. In this paper, we propose a novel multi-perspective cross-lingual neural framework for code–text matching, inspired in part by a previou...
c52407ecf5f8336d34ff9ea917b841f1
2,020
[ "the ability to match pieces of code to their corresponding natural language descriptions and vice versa is fundamental for natural language search interfaces to software repositories .", "in this paper , we propose a novel multi - perspective cross - lingual neural framework for code – text matching , inspired i...
[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 31 ] }, { "text": "multi - perspective cross - lingual neural framewo...
[ "the", "ability", "to", "match", "pieces", "of", "code", "to", "their", "corresponding", "natural", "language", "descriptions", "and", "vice", "versa", "is", "fundamental", "for", "natural", "language", "search", "interfaces", "to", "software", "repositories", "."...
ACL
Reports of personal experiences and stories in argumentation: datasets and analysis
Reports of personal experiences or stories can play a crucial role in argumentation, as they represent an immediate and (often) relatable way to back up one’s position with respect to a given topic. They are easy to understand and increase empathy: this makes them powerful in argumentation. The impact of personal repor...
1fdb6a91d43ced64d916f25779bb40c0
2,022
[ "reports of personal experiences or stories can play a crucial role in argumentation , as they represent an immediate and ( often ) relatable way to back up one ’ s position with respect to a given topic .", "they are easy to understand and increase empathy : this makes them powerful in argumentation .", "the i...
[ { "event_type": "WKS", "arguments": [ { "text": "our", "nugget_type": "OG", "argument_type": "Researcher", "tokens": [ "our" ], "offsets": [ 92 ] }, { "text": "robust classifiers", "nugget_type": "A...
[ "reports", "of", "personal", "experiences", "or", "stories", "can", "play", "a", "crucial", "role", "in", "argumentation", ",", "as", "they", "represent", "an", "immediate", "and", "(", "often", ")", "relatable", "way", "to", "back", "up", "one", "’", "s",...
ACL
FIBER: Fill-in-the-Blanks as a Challenging Video Understanding Evaluation Framework
We propose fill-in-the-blanks as a video understanding evaluation framework and introduce FIBER – a novel dataset consisting of 28,000 videos and descriptions in support of this evaluation framework. The fill-in-the-blanks setting tests a model’s understanding of a video by requiring it to predict a masked noun phrase ...
4ef74fc71e61b1c5b9b124a6adacd878
2,022
[ "we propose fill - in - the - blanks as a video understanding evaluation framework and introduce fiber – a novel dataset consisting of 28 , 000 videos and descriptions in support of this evaluation framework .", "the fill - in - the - blanks setting tests a model ’ s understanding of a video by requiring it to pr...
[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 0 ] }, { "text": "video understanding evaluation framework", ...
[ "we", "propose", "fill", "-", "in", "-", "the", "-", "blanks", "as", "a", "video", "understanding", "evaluation", "framework", "and", "introduce", "fiber", "–", "a", "novel", "dataset", "consisting", "of", "28", ",", "000", "videos", "and", "descriptions", ...
ACL
HeterMPC: A Heterogeneous Graph Neural Network for Response Generation in Multi-Party Conversations
Recently, various response generation models for two-party conversations have achieved impressive improvements, but less effort has been paid to multi-party conversations (MPCs) which are more practical and complicated. Compared with a two-party conversation where a dialogue context is a sequence of utterances, buildin...
cb06f8b44bec4044aef365678d970940
2,022
[ "recently , various response generation models for two - party conversations have achieved impressive improvements , but less effort has been paid to multi - party conversations ( mpcs ) which are more practical and complicated .", "compared with a two - party conversation where a dialogue context is a sequence o...
[ { "event_type": "ITT", "arguments": [ { "text": "response generation models", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "response", "generation", "models" ], "offsets": [ 3, 4, ...
[ "recently", ",", "various", "response", "generation", "models", "for", "two", "-", "party", "conversations", "have", "achieved", "impressive", "improvements", ",", "but", "less", "effort", "has", "been", "paid", "to", "multi", "-", "party", "conversations", "(",...
ACL
TopWORDS-Seg: Simultaneous Text Segmentation and Word Discovery for Open-Domain Chinese Texts via Bayesian Inference
Processing open-domain Chinese texts has been a critical bottleneck in computational linguistics for decades, partially because text segmentation and word discovery often entangle with each other in this challenging scenario. No existing methods yet can achieve effective text segmentation and word discovery simultaneou...
25236bf130ccbbc55d4f5574c2383f7d
2,022
[ "processing open - domain chinese texts has been a critical bottleneck in computational linguistics for decades , partially because text segmentation and word discovery often entangle with each other in this challenging scenario .", "no existing methods yet can achieve effective text segmentation and word discove...
[ { "event_type": "ITT", "arguments": [ { "text": "open - domain chinese texts", "nugget_type": "FEA", "argument_type": "Target", "tokens": [ "open", "-", "domain", "chinese", "texts" ], "offsets": [ ...
[ "processing", "open", "-", "domain", "chinese", "texts", "has", "been", "a", "critical", "bottleneck", "in", "computational", "linguistics", "for", "decades", ",", "partially", "because", "text", "segmentation", "and", "word", "discovery", "often", "entangle", "wi...
ACL
Overestimation of Syntactic Representation in Neural Language Models
With the advent of powerful neural language models over the last few years, research attention has increasingly focused on what aspects of language they represent that make them so successful. Several testing methodologies have been developed to probe models’ syntactic representations. One popular method for determinin...
04445403b08ebbe1ea95c9ff3311fc89
2,020
[ "with the advent of powerful neural language models over the last few years , research attention has increasingly focused on what aspects of language they represent that make them so successful .", "several testing methodologies have been developed to probe models ’ syntactic representations .", "one popular me...
[ { "event_type": "ITT", "arguments": [ { "text": "neural language models", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "neural", "language", "models" ], "offsets": [ 5, 6, 7 ...
[ "with", "the", "advent", "of", "powerful", "neural", "language", "models", "over", "the", "last", "few", "years", ",", "research", "attention", "has", "increasingly", "focused", "on", "what", "aspects", "of", "language", "they", "represent", "that", "make", "t...
ACL
Lipschitz Constrained Parameter Initialization for Deep Transformers
The Transformer translation model employs residual connection and layer normalization to ease the optimization difficulties caused by its multi-layer encoder/decoder structure. Previous research shows that even with residual connection and layer normalization, deep Transformers still have difficulty in training, and pa...
10003fbb3bbbbd32bcc94760868e1b0f
2,020
[ "the transformer translation model employs residual connection and layer normalization to ease the optimization difficulties caused by its multi - layer encoder / decoder structure .", "previous research shows that even with residual connection and layer normalization , deep transformers still have difficulty in ...
[ { "event_type": "FAC", "arguments": [ { "text": "parameter initialization method", "nugget_type": "APP", "argument_type": "Subject", "tokens": [ "parameter", "initialization", "method" ], "offsets": [ 117, ...
[ "the", "transformer", "translation", "model", "employs", "residual", "connection", "and", "layer", "normalization", "to", "ease", "the", "optimization", "difficulties", "caused", "by", "its", "multi", "-", "layer", "encoder", "/", "decoder", "structure", ".", "pre...
ACL
Multi-Granularity Structural Knowledge Distillation for Language Model Compression
Transferring the knowledge to a small model through distillation has raised great interest in recent years. Prevailing methods transfer the knowledge derived from mono-granularity language units (e.g., token-level or sample-level), which is not enough to represent the rich semantics of a text and may lose some vital kn...
3d9c10478013790a0d8378aab134a404
2,022
[ "transferring the knowledge to a small model through distillation has raised great interest in recent years .", "prevailing methods transfer the knowledge derived from mono - granularity language units ( e . g . , token - level or sample - level ) , which is not enough to represent the rich semantics of a text an...
[ { "event_type": "RWS", "arguments": [ { "text": "knowledge", "nugget_type": "FEA", "argument_type": "TriedComponent", "tokens": [ "knowledge" ], "offsets": [ 2 ] }, { "text": "distillation", "nugget...
[ "transferring", "the", "knowledge", "to", "a", "small", "model", "through", "distillation", "has", "raised", "great", "interest", "in", "recent", "years", ".", "prevailing", "methods", "transfer", "the", "knowledge", "derived", "from", "mono", "-", "granularity", ...
ACL
Beyond Offline Mapping: Learning Cross-lingual Word Embeddings through Context Anchoring
Recent research on cross-lingual word embeddings has been dominated by unsupervised mapping approaches that align monolingual embeddings. Such methods critically rely on those embeddings having a similar structure, but it was recently shown that the separate training in different languages causes departures from this a...
f7171131307b5f72a3e17ce88cb2b879
2,021
[ "recent research on cross - lingual word embeddings has been dominated by unsupervised mapping approaches that align monolingual embeddings .", "such methods critically rely on those embeddings having a similar structure , but it was recently shown that the separate training in different languages causes departur...
[ { "event_type": "ITT", "arguments": [ { "text": "cross - lingual word embeddings", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "cross", "-", "lingual", "word", "embeddings" ], "offsets": ...
[ "recent", "research", "on", "cross", "-", "lingual", "word", "embeddings", "has", "been", "dominated", "by", "unsupervised", "mapping", "approaches", "that", "align", "monolingual", "embeddings", ".", "such", "methods", "critically", "rely", "on", "those", "embedd...
ACL
CAMERO: Consistency Regularized Ensemble of Perturbed Language Models with Weight Sharing
Model ensemble is a popular approach to produce a low-variance and well-generalized model. However, it induces large memory and inference costs, which is often not affordable for real-world deployment. Existing work has resorted to sharing weights among models. However, when increasing the proportion of the shared weig...
c21be2d2c9cf955f86e0fbb1ce918e1b
2,022
[ "model ensemble is a popular approach to produce a low - variance and well - generalized model .", "however , it induces large memory and inference costs , which is often not affordable for real - world deployment .", "existing work has resorted to sharing weights among models .", "however , when increasing t...
[ { "event_type": "MDS", "arguments": [ { "text": "prediction consistency regularizer", "nugget_type": "MOD", "argument_type": "TriedComponent", "tokens": [ "prediction", "consistency", "regularizer" ], "offsets": [ ...
[ "model", "ensemble", "is", "a", "popular", "approach", "to", "produce", "a", "low", "-", "variance", "and", "well", "-", "generalized", "model", ".", "however", ",", "it", "induces", "large", "memory", "and", "inference", "costs", ",", "which", "is", "ofte...
ACL
Effective Estimation of Deep Generative Language Models
Advances in variational inference enable parameterisation of probabilistic models by deep neural networks. This combines the statistical transparency of the probabilistic modelling framework with the representational power of deep learning. Yet, due to a problem known as posterior collapse, it is difficult to estimate ...
6e803a5c3601838b5832804d9180e868
2,020
[ "advances in variational inference enable parameterisation of probabilistic models by deep neural networks .", "this combines the statistical transparency of the probabilistic modelling framework with the representational power of deep learning .", "yet , due to a problem known as posterior collapse , it is dif...
[ { "event_type": "ITT", "arguments": [ { "text": "parameterisation of probabilistic models", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "parameterisation", "of", "probabilistic", "models" ], "offse...
[ "advances", "in", "variational", "inference", "enable", "parameterisation", "of", "probabilistic", "models", "by", "deep", "neural", "networks", ".", "this", "combines", "the", "statistical", "transparency", "of", "the", "probabilistic", "modelling", "framework", "wit...
ACL
MELM: Data Augmentation with Masked Entity Language Modeling for Low-Resource NER
Data augmentation is an effective solution to data scarcity in low-resource scenarios. However, when applied to token-level tasks such as NER, data augmentation methods often suffer from token-label misalignment, which leads to unsatsifactory performance. In this work, we propose Masked Entity Language Modeling (MELM) ...
1573320df13703afc93e3cf31172bd81
2,022
[ "data augmentation is an effective solution to data scarcity in low - resource scenarios .", "however , when applied to token - level tasks such as ner , data augmentation methods often suffer from token - label misalignment , which leads to unsatsifactory performance .", "in this work , we propose masked entit...
[ { "event_type": "ITT", "arguments": [ { "text": "data augmentation", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "data", "augmentation" ], "offsets": [ 0, 1 ] } ], "trigger": ...
[ "data", "augmentation", "is", "an", "effective", "solution", "to", "data", "scarcity", "in", "low", "-", "resource", "scenarios", ".", "however", ",", "when", "applied", "to", "token", "-", "level", "tasks", "such", "as", "ner", ",", "data", "augmentation", ...
ACL
Fully Hyperbolic Neural Networks
Hyperbolic neural networks have shown great potential for modeling complex data. However, existing hyperbolic networks are not completely hyperbolic, as they encode features in the hyperbolic space yet formalize most of their operations in the tangent space (a Euclidean subspace) at the origin of the hyperbolic model. ...
66ce7993b51b2f48b25cf8ed72e7ab14
2,022
[ "hyperbolic neural networks have shown great potential for modeling complex data .", "however , existing hyperbolic networks are not completely hyperbolic , as they encode features in the hyperbolic space yet formalize most of their operations in the tangent space ( a euclidean subspace ) at the origin of the hyp...
[ { "event_type": "ITT", "arguments": [ { "text": "complex data", "nugget_type": "DST", "argument_type": "Target", "tokens": [ "complex", "data" ], "offsets": [ 9, 10 ] } ], "trigger": { "...
[ "hyperbolic", "neural", "networks", "have", "shown", "great", "potential", "for", "modeling", "complex", "data", ".", "however", ",", "existing", "hyperbolic", "networks", "are", "not", "completely", "hyperbolic", ",", "as", "they", "encode", "features", "in", "...
ACL
Learning Architectures from an Extended Search Space for Language Modeling
Neural architecture search (NAS) has advanced significantly in recent years but most NAS systems restrict search to learning architectures of a recurrent or convolutional cell. In this paper, we extend the search space of NAS. In particular, we present a general approach to learn both intra-cell and inter-cell architec...
e94325f07738bf8faf00398c9bcb4427
2,020
[ "neural architecture search ( nas ) has advanced significantly in recent years but most nas systems restrict search to learning architectures of a recurrent or convolutional cell .", "in this paper , we extend the search space of nas .", "in particular , we present a general approach to learn both intra - cell ...
[ { "event_type": "ITT", "arguments": [ { "text": "neural architecture search", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "neural", "architecture", "search" ], "offsets": [ 0, 1, ...
[ "neural", "architecture", "search", "(", "nas", ")", "has", "advanced", "significantly", "in", "recent", "years", "but", "most", "nas", "systems", "restrict", "search", "to", "learning", "architectures", "of", "a", "recurrent", "or", "convolutional", "cell", "."...
ACL
A Corpus for Modeling User and Language Effects in Argumentation on Online Debating
Existing argumentation datasets have succeeded in allowing researchers to develop computational methods for analyzing the content, structure and linguistic features of argumentative text. They have been much less successful in fostering studies of the effect of “user” traits — characteristics and beliefs of the partici...
adbc900a63b1e20d3f24412e96816efe
2,019
[ "existing argumentation datasets have succeeded in allowing researchers to develop computational methods for analyzing the content , structure and linguistic features of argumentative text .", "they have been much less successful in fostering studies of the effect of “ user ” traits — characteristics and beliefs ...
[ { "event_type": "ITT", "arguments": [ { "text": "argumentation datasets", "nugget_type": "DST", "argument_type": "Target", "tokens": [ "argumentation", "datasets" ], "offsets": [ 1, 2 ] } ], "...
[ "existing", "argumentation", "datasets", "have", "succeeded", "in", "allowing", "researchers", "to", "develop", "computational", "methods", "for", "analyzing", "the", "content", ",", "structure", "and", "linguistic", "features", "of", "argumentative", "text", ".", "...
ACL
Leveraging Meta Information in Short Text Aggregation
Short texts such as tweets often contain insufficient word co-occurrence information for training conventional topic models. To deal with the insufficiency, we propose a generative model that aggregates short texts into clusters by leveraging the associated meta information. Our model can generate more interpretable to...
50e61d46a43821bf92c6a2e6e5ad27f3
2,019
[ "short texts such as tweets often contain insufficient word co - occurrence information for training conventional topic models .", "to deal with the insufficiency , we propose a generative model that aggregates short texts into clusters by leveraging the associated meta information .", "our model can generate m...
[ { "event_type": "RWF", "arguments": [ { "text": "short texts", "nugget_type": "DST", "argument_type": "Concern", "tokens": [ "short", "texts" ], "offsets": [ 0, 1 ] }, { "text": "insuffi...
[ "short", "texts", "such", "as", "tweets", "often", "contain", "insufficient", "word", "co", "-", "occurrence", "information", "for", "training", "conventional", "topic", "models", ".", "to", "deal", "with", "the", "insufficiency", ",", "we", "propose", "a", "g...
ACL
Detecting Perceived Emotions in Hurricane Disasters
Natural disasters (e.g., hurricanes) affect millions of people each year, causing widespread destruction in their wake. People have recently taken to social media websites (e.g., Twitter) to share their sentiments and feelings with the larger community. Consequently, these platforms have become instrumental in understa...
2dd630e8f83ca0f370ff9084b1a4c0d0
2,020
[ "natural disasters ( e . g . , hurricanes ) affect millions of people each year , causing widespread destruction in their wake .", "people have recently taken to social media websites ( e . g . , twitter ) to share their sentiments and feelings with the larger community .", "consequently , these platforms have ...
[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 70 ] }, { "text": "hurricaneemo", "nugget_type": "DST", ...
[ "natural", "disasters", "(", "e", ".", "g", ".", ",", "hurricanes", ")", "affect", "millions", "of", "people", "each", "year", ",", "causing", "widespread", "destruction", "in", "their", "wake", ".", "people", "have", "recently", "taken", "to", "social", "...
ACL
Benefits of Intermediate Annotations in Reading Comprehension
Complex compositional reading comprehension datasets require performing latent sequential decisions that are learned via supervision from the final answer. A large combinatorial space of possible decision paths that result in the same answer, compounded by the lack of intermediate supervision to help choose the right p...
49ce460bcdcc98f8fcdc54b604d2de7b
2,020
[ "complex compositional reading comprehension datasets require performing latent sequential decisions that are learned via supervision from the final answer .", "a large combinatorial space of possible decision paths that result in the same answer , compounded by the lack of intermediate supervision to help choose...
[ { "event_type": "ITT", "arguments": [ { "text": "complex compositional reading comprehension datasets", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "complex", "compositional", "reading", "comprehension", ...
[ "complex", "compositional", "reading", "comprehension", "datasets", "require", "performing", "latent", "sequential", "decisions", "that", "are", "learned", "via", "supervision", "from", "the", "final", "answer", ".", "a", "large", "combinatorial", "space", "of", "po...
ACL
Modeling Temporal-Modal Entity Graph for Procedural Multimodal Machine Comprehension
Procedural Multimodal Documents (PMDs) organize textual instructions and corresponding images step by step. Comprehending PMDs and inducing their representations for the downstream reasoning tasks is designated as Procedural MultiModal Machine Comprehension (M3C). In this study, we approach Procedural M3C at a fine-gra...
2ba26475c6052d46e2c41af143408ff2
2,022
[ "procedural multimodal documents ( pmds ) organize textual instructions and corresponding images step by step .", "comprehending pmds and inducing their representations for the downstream reasoning tasks is designated as procedural multimodal machine comprehension ( m3c ) .", "in this study , we approach proced...
[ { "event_type": "RWS", "arguments": [ { "text": "procedural multimodal documents", "nugget_type": "APP", "argument_type": "Subject", "tokens": [ "procedural", "multimodal", "documents" ], "offsets": [ 0, ...
[ "procedural", "multimodal", "documents", "(", "pmds", ")", "organize", "textual", "instructions", "and", "corresponding", "images", "step", "by", "step", ".", "comprehending", "pmds", "and", "inducing", "their", "representations", "for", "the", "downstream", "reason...
ACL
Reducing Position Bias in Simultaneous Machine Translation with Length-Aware Framework
Simultaneous machine translation (SiMT) starts translating while receiving the streaming source inputs, and hence the source sentence is always incomplete during translating. Different from the full-sentence MT using the conventional seq-to-seq architecture, SiMT often applies prefix-to-prefix architecture, which force...
3c64d537b1ac5f6ae0eb0294513f1546
2,022
[ "simultaneous machine translation ( simt ) starts translating while receiving the streaming source inputs , and hence the source sentence is always incomplete during translating .", "different from the full - sentence mt using the conventional seq - to - seq architecture , simt often applies prefix - to - prefix ...
[ { "event_type": "ITT", "arguments": [ { "text": "simt", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "simt" ], "offsets": [ 151 ] } ], "trigger": { "text": "starts", "tokens": [ ...
[ "simultaneous", "machine", "translation", "(", "simt", ")", "starts", "translating", "while", "receiving", "the", "streaming", "source", "inputs", ",", "and", "hence", "the", "source", "sentence", "is", "always", "incomplete", "during", "translating", ".", "differ...
ACL
Compact Token Representations with Contextual Quantization for Efficient Document Re-ranking
Transformer based re-ranking models can achieve high search relevance through context- aware soft matching of query tokens with document tokens. To alleviate runtime complexity of such inference, previous work has adopted a late interaction architecture with pre-computed contextual token representations at the cost of ...
5e2c2d750af8ffb94e7f2a83b034337b
2,022
[ "transformer based re - ranking models can achieve high search relevance through context - aware soft matching of query tokens with document tokens .", "to alleviate runtime complexity of such inference , previous work has adopted a late interaction architecture with pre - computed contextual token representation...
[ { "event_type": "RWS", "arguments": [ { "text": "late interaction architecture", "nugget_type": "APP", "argument_type": "BaseComponent", "tokens": [ "late", "interaction", "architecture" ], "offsets": [ 37, ...
[ "transformer", "based", "re", "-", "ranking", "models", "can", "achieve", "high", "search", "relevance", "through", "context", "-", "aware", "soft", "matching", "of", "query", "tokens", "with", "document", "tokens", ".", "to", "alleviate", "runtime", "complexity...
ACL
So Different Yet So Alike! Constrained Unsupervised Text Style Transfer
Automatic transfer of text between domains has become popular in recent times. One of its aims is to preserve the semantic content while adapting to the target domain. However, it does not explicitly maintain other attributes between the source and translated text: e.g., text length and descriptiveness. Maintaining con...
e6d64453528e8588dc9571de84676b9c
2,022
[ "automatic transfer of text between domains has become popular in recent times .", "one of its aims is to preserve the semantic content while adapting to the target domain .", "however , it does not explicitly maintain other attributes between the source and translated text : e . g . , text length and descripti...
[ { "event_type": "ITT", "arguments": [ { "text": "automatic transfer of text", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "automatic", "transfer", "of", "text" ], "offsets": [ 0, ...
[ "automatic", "transfer", "of", "text", "between", "domains", "has", "become", "popular", "in", "recent", "times", ".", "one", "of", "its", "aims", "is", "to", "preserve", "the", "semantic", "content", "while", "adapting", "to", "the", "target", "domain", "."...
ACL
Neural Legal Judgment Prediction in English
Legal judgment prediction is the task of automatically predicting the outcome of a court case, given a text describing the case’s facts. Previous work on using neural models for this task has focused on Chinese; only feature-based models (e.g., using bags of words and topics) have been considered in English. We release...
d6f45c5e1c4c76db5f014fc5240d4203
2,019
[ "legal judgment prediction is the task of automatically predicting the outcome of a court case , given a text describing the case ’ s facts .", "previous work on using neural models for this task has focused on chinese ; only feature - based models ( e . g . , using bags of words and topics ) have been considered...
[ { "event_type": "ITT", "arguments": [ { "text": "legal judgment prediction", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "legal", "judgment", "prediction" ], "offsets": [ 0, 1, ...
[ "legal", "judgment", "prediction", "is", "the", "task", "of", "automatically", "predicting", "the", "outcome", "of", "a", "court", "case", ",", "given", "a", "text", "describing", "the", "case", "’", "s", "facts", ".", "previous", "work", "on", "using", "n...
ACL
A Cross-Sentence Latent Variable Model for Semi-Supervised Text Sequence Matching
We present a latent variable model for predicting the relationship between a pair of text sequences. Unlike previous auto-encoding–based approaches that consider each sequence separately, our proposed framework utilizes both sequences within a single model by generating a sequence that has a given relationship with a s...
e773dc0cbb85006aabc537e2bdea183b
2,019
[ "we present a latent variable model for predicting the relationship between a pair of text sequences .", "unlike previous auto - encoding – based approaches that consider each sequence separately , our proposed framework utilizes both sequences within a single model by generating a sequence that has a given relat...
[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 0 ] }, { "text": "latent variable model", "nugget_type": "APP...
[ "we", "present", "a", "latent", "variable", "model", "for", "predicting", "the", "relationship", "between", "a", "pair", "of", "text", "sequences", ".", "unlike", "previous", "auto", "-", "encoding", "–", "based", "approaches", "that", "consider", "each", "seq...
ACL
Exploring Author Context for Detecting Intended vs Perceived Sarcasm
We investigate the impact of using author context on textual sarcasm detection. We define author context as the embedded representation of their historical posts on Twitter and suggest neural models that extract these representations. We experiment with two tweet datasets, one labelled manually for sarcasm, and the oth...
8701994001de9048844b7c855bc1d558
2,019
[ "we investigate the impact of using author context on textual sarcasm detection .", "we define author context as the embedded representation of their historical posts on twitter and suggest neural models that extract these representations .", "we experiment with two tweet datasets , one labelled manually for sa...
[ { "event_type": "WKS", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Researcher", "tokens": [ "we" ], "offsets": [ 0 ] }, { "text": "impact of using author context on textual sarcasm...
[ "we", "investigate", "the", "impact", "of", "using", "author", "context", "on", "textual", "sarcasm", "detection", ".", "we", "define", "author", "context", "as", "the", "embedded", "representation", "of", "their", "historical", "posts", "on", "twitter", "and", ...
ACL
A Deep Reinforced Sequence-to-Set Model for Multi-Label Classification
Multi-label classification (MLC) aims to predict a set of labels for a given instance. Based on a pre-defined label order, the sequence-to-sequence (Seq2Seq) model trained via maximum likelihood estimation method has been successfully applied to the MLC task and shows powerful ability to capture high-order correlations...
ae14e4dda6ddc78edd8b83be9d5b0165
2,019
[ "multi - label classification ( mlc ) aims to predict a set of labels for a given instance .", "based on a pre - defined label order , the sequence - to - sequence ( seq2seq ) model trained via maximum likelihood estimation method has been successfully applied to the mlc task and shows powerful ability to capture...
[ { "event_type": "ITT", "arguments": [ { "text": "multi - label classification", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "multi", "-", "label", "classification" ], "offsets": [ 0, ...
[ "multi", "-", "label", "classification", "(", "mlc", ")", "aims", "to", "predict", "a", "set", "of", "labels", "for", "a", "given", "instance", ".", "based", "on", "a", "pre", "-", "defined", "label", "order", ",", "the", "sequence", "-", "to", "-", ...
ACL
Mismatch between Multi-turn Dialogue and its Evaluation Metric in Dialogue State Tracking
Dialogue state tracking (DST) aims to extract essential information from multi-turn dialog situations and take appropriate actions. A belief state, one of the core pieces of information, refers to the subject and its specific content, and appears in the form of domain-slot-value. The trained model predicts “accumulated...
9eaffb78d7b941442a2994c52ae2595a
2,022
[ "dialogue state tracking ( dst ) aims to extract essential information from multi - turn dialog situations and take appropriate actions .", "a belief state , one of the core pieces of information , refers to the subject and its specific content , and appears in the form of domain - slot - value .", "the trained...
[ { "event_type": "ITT", "arguments": [ { "text": "dialogue state tracking", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "dialogue", "state", "tracking" ], "offsets": [ 0, 1, 2 ...
[ "dialogue", "state", "tracking", "(", "dst", ")", "aims", "to", "extract", "essential", "information", "from", "multi", "-", "turn", "dialog", "situations", "and", "take", "appropriate", "actions", ".", "a", "belief", "state", ",", "one", "of", "the", "core"...
ACL
Leveraging Similar Users for Personalized Language Modeling with Limited Data
Personalized language models are designed and trained to capture language patterns specific to individual users. This makes them more accurate at predicting what a user will write. However, when a new user joins a platform and not enough text is available, it is harder to build effective personalized language models. W...
bca6adf935b57aa3bb3d4b2b9f60199b
2,022
[ "personalized language models are designed and trained to capture language patterns specific to individual users .", "this makes them more accurate at predicting what a user will write .", "however , when a new user joins a platform and not enough text is available , it is harder to build effective personalized...
[ { "event_type": "ITT", "arguments": [ { "text": "personalized language models", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "personalized", "language", "models" ], "offsets": [ 0, 1, ...
[ "personalized", "language", "models", "are", "designed", "and", "trained", "to", "capture", "language", "patterns", "specific", "to", "individual", "users", ".", "this", "makes", "them", "more", "accurate", "at", "predicting", "what", "a", "user", "will", "write...
ACL
How to Adapt Your Pretrained Multilingual Model to 1600 Languages
Pretrained multilingual models (PMMs) enable zero-shot learning via cross-lingual transfer, performing best for languages seen during pretraining. While methods exist to improve performance for unseen languages, they have almost exclusively been evaluated using amounts of raw text only available for a small fraction of...
e4e97fa9531e04d9c026e3f35fb1fb53
2,021
[ "pretrained multilingual models ( pmms ) enable zero - shot learning via cross - lingual transfer , performing best for languages seen during pretraining .", "while methods exist to improve performance for unseen languages , they have almost exclusively been evaluated using amounts of raw text only available for ...
[ { "event_type": "ITT", "arguments": [ { "text": "pretrained multilingual models", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "pretrained", "multilingual", "models" ], "offsets": [ 0, 1, ...
[ "pretrained", "multilingual", "models", "(", "pmms", ")", "enable", "zero", "-", "shot", "learning", "via", "cross", "-", "lingual", "transfer", ",", "performing", "best", "for", "languages", "seen", "during", "pretraining", ".", "while", "methods", "exist", "...
ACL
Unsupervised Extractive Summarization-Based Representations for Accurate and Explainable Collaborative Filtering
We pioneer the first extractive summarization-based collaborative filtering model called ESCOFILT. Our proposed model specifically produces extractive summaries for each item and user. Unlike other types of explanations, summary-level explanations closely resemble real-life explanations. The strength of ESCOFILT lies i...
c6c6186a840704417999b9e5a32ba87a
2,021
[ "we pioneer the first extractive summarization - based collaborative filtering model called escofilt .", "our proposed model specifically produces extractive summaries for each item and user .", "unlike other types of explanations , summary - level explanations closely resemble real - life explanations .", "t...
[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 0 ] }, { "text": "escofilt", "nugget_type": "APP", "a...
[ "we", "pioneer", "the", "first", "extractive", "summarization", "-", "based", "collaborative", "filtering", "model", "called", "escofilt", ".", "our", "proposed", "model", "specifically", "produces", "extractive", "summaries", "for", "each", "item", "and", "user", ...
ACL
EarlyBERT: Efficient BERT Training via Early-bird Lottery Tickets
Heavily overparameterized language models such as BERT, XLNet and T5 have achieved impressive success in many NLP tasks. However, their high model complexity requires enormous computation resources and extremely long training time for both pre-training and fine-tuning. Many works have studied model compression on large...
7460a7800f720a95eb642d4f87ed56e3
2,021
[ "heavily overparameterized language models such as bert , xlnet and t5 have achieved impressive success in many nlp tasks .", "however , their high model complexity requires enormous computation resources and extremely long training time for both pre - training and fine - tuning .", "many works have studied mod...
[ { "event_type": "ITT", "arguments": [ { "text": "heavily overparameterized language models", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "heavily", "overparameterized", "language", "models" ], "off...
[ "heavily", "overparameterized", "language", "models", "such", "as", "bert", ",", "xlnet", "and", "t5", "have", "achieved", "impressive", "success", "in", "many", "nlp", "tasks", ".", "however", ",", "their", "high", "model", "complexity", "requires", "enormous",...
ACL
Semi-Supervised Semantic Dependency Parsing Using CRF Autoencoders
Semantic dependency parsing, which aims to find rich bi-lexical relationships, allows words to have multiple dependency heads, resulting in graph-structured representations. We propose an approach to semi-supervised learning of semantic dependency parsers based on the CRF autoencoder framework. Our encoder is a discrim...
d53a5a5071286d973514b9e427f4b520
2,020
[ "semantic dependency parsing , which aims to find rich bi - lexical relationships , allows words to have multiple dependency heads , resulting in graph - structured representations .", "we propose an approach to semi - supervised learning of semantic dependency parsers based on the crf autoencoder framework .", ...
[ { "event_type": "ITT", "arguments": [ { "text": "semantic dependency parsing", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "semantic", "dependency", "parsing" ], "offsets": [ 0, 1, ...
[ "semantic", "dependency", "parsing", ",", "which", "aims", "to", "find", "rich", "bi", "-", "lexical", "relationships", ",", "allows", "words", "to", "have", "multiple", "dependency", "heads", ",", "resulting", "in", "graph", "-", "structured", "representations"...
ACL
How Did This Get Funded?! Automatically Identifying Quirky Scientific Achievements
Humor is an important social phenomenon, serving complex social and psychological functions. However, despite being studied for millennia humor is computationally not well understood, often considered an AI-complete problem. In this work, we introduce a novel setting in humor mining: automatically detecting funny and u...
9e9a8561a44bc51b89559c0ad736dce3
2,021
[ "humor is an important social phenomenon , serving complex social and psychological functions .", "however , despite being studied for millennia humor is computationally not well understood , often considered an ai - complete problem .", "in this work , we introduce a novel setting in humor mining : automatical...
[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 40 ] }, { "text": "setting in humor mining", "nugget_type": "...
[ "humor", "is", "an", "important", "social", "phenomenon", ",", "serving", "complex", "social", "and", "psychological", "functions", ".", "however", ",", "despite", "being", "studied", "for", "millennia", "humor", "is", "computationally", "not", "well", "understood...
ACL
PARE: A Simple and Strong Baseline for Monolingual and Multilingual Distantly Supervised Relation Extraction
Neural models for distantly supervised relation extraction (DS-RE) encode each sentence in an entity-pair bag separately. These are then aggregated for bag-level relation prediction. Since, at encoding time, these approaches do not allow information to flow from other sentences in the bag, we believe that they do not u...
896bd48bcb383a54604d82cc357b668f
2,022
[ "neural models for distantly supervised relation extraction ( ds - re ) encode each sentence in an entity - pair bag separately .", "these are then aggregated for bag - level relation prediction .", "since , at encoding time , these approaches do not allow information to flow from other sentences in the bag , w...
[ { "event_type": "RWS", "arguments": [ { "text": "distantly supervised relation extraction", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "distantly", "supervised", "relation", "extraction" ], "offse...
[ "neural", "models", "for", "distantly", "supervised", "relation", "extraction", "(", "ds", "-", "re", ")", "encode", "each", "sentence", "in", "an", "entity", "-", "pair", "bag", "separately", ".", "these", "are", "then", "aggregated", "for", "bag", "-", "...
ACL
Flow-Adapter Architecture for Unsupervised Machine Translation
In this work, we propose a flow-adapter architecture for unsupervised NMT. It leverages normalizing flows to explicitly model the distributions of sentence-level latent representations, which are subsequently used in conjunction with the attention mechanism for the translation task. The primary novelties of our model a...
eed953c88714d4fe6759bbaec5c6aa8a
2,022
[ "in this work , we propose a flow - adapter architecture for unsupervised nmt .", "it leverages normalizing flows to explicitly model the distributions of sentence - level latent representations , which are subsequently used in conjunction with the attention mechanism for the translation task .", "the primary n...
[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 4 ] }, { "text": "flow - adapter architecture", "nugget_type"...
[ "in", "this", "work", ",", "we", "propose", "a", "flow", "-", "adapter", "architecture", "for", "unsupervised", "nmt", ".", "it", "leverages", "normalizing", "flows", "to", "explicitly", "model", "the", "distributions", "of", "sentence", "-", "level", "latent"...
ACL
Toward Annotator Group Bias in Crowdsourcing
Crowdsourcing has emerged as a popular approach for collecting annotated data to train supervised machine learning models. However, annotator bias can lead to defective annotations. Though there are a few works investigating individual annotator bias, the group effects in annotators are largely overlooked. In this work...
f64d64a7cf39cafdb39d2d507f6301d0
2,022
[ "crowdsourcing has emerged as a popular approach for collecting annotated data to train supervised machine learning models .", "however , annotator bias can lead to defective annotations .", "though there are a few works investigating individual annotator bias , the group effects in annotators are largely overl...
[ { "event_type": "ITT", "arguments": [ { "text": "crowdsourcing", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "crowdsourcing" ], "offsets": [ 0 ] } ], "trigger": { "text": "emerged", "...
[ "crowdsourcing", "has", "emerged", "as", "a", "popular", "approach", "for", "collecting", "annotated", "data", "to", "train", "supervised", "machine", "learning", "models", ".", "however", ",", "annotator", "bias", "can", "lead", "to", "defective", "annotations", ...
ACL
W-RST: Towards a Weighted RST-style Discourse Framework
Aiming for a better integration of data-driven and linguistically-inspired approaches, we explore whether RST Nuclearity, assigning a binary assessment of importance between text segments, can be replaced by automatically generated, real-valued scores, in what we call a Weighted-RST framework. In particular, we find th...
a4d5254f815e7ade9ec9412783a708aa
2,021
[ "aiming for a better integration of data - driven and linguistically - inspired approaches , we explore whether rst nuclearity , assigning a binary assessment of importance between text segments , can be replaced by automatically generated , real - valued scores , in what we call a weighted - rst framework .", "i...
[ { "event_type": "FIN", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Finder", "tokens": [ "we" ], "offsets": [ 56 ] }, { "text": "benefit", "nugget_type": "E-CMP", "a...
[ "aiming", "for", "a", "better", "integration", "of", "data", "-", "driven", "and", "linguistically", "-", "inspired", "approaches", ",", "we", "explore", "whether", "rst", "nuclearity", ",", "assigning", "a", "binary", "assessment", "of", "importance", "between"...
ACL
Instantaneous Grammatical Error Correction with Shallow Aggressive Decoding
In this paper, we propose Shallow Aggressive Decoding (SAD) to improve the online inference efficiency of the Transformer for instantaneous Grammatical Error Correction (GEC). SAD optimizes the online inference efficiency for GEC by two innovations: 1) it aggressively decodes as many tokens as possible in parallel inst...
ae52da18eee3cd47ff2e3d91407fe8cc
2,021
[ "in this paper , we propose shallow aggressive decoding ( sad ) to improve the online inference efficiency of the transformer for instantaneous grammatical error correction ( gec ) .", "sad optimizes the online inference efficiency for gec by two innovations : 1 ) it aggressively decodes as many tokens as possibl...
[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 4 ] }, { "text": "shallow aggressive decoding", "nugget_type"...
[ "in", "this", "paper", ",", "we", "propose", "shallow", "aggressive", "decoding", "(", "sad", ")", "to", "improve", "the", "online", "inference", "efficiency", "of", "the", "transformer", "for", "instantaneous", "grammatical", "error", "correction", "(", "gec", ...
ACL
StereoRel: Relational Triple Extraction from a Stereoscopic Perspective
Relational triple extraction is critical to understanding massive text corpora and constructing large-scale knowledge graph, which has attracted increasing research interest. However, existing studies still face some challenging issues, including information loss, error propagation and ignoring the interaction between ...
4e174fb9de813efa12f02eabcb9f2591
2,021
[ "relational triple extraction is critical to understanding massive text corpora and constructing large - scale knowledge graph , which has attracted increasing research interest .", "however , existing studies still face some challenging issues , including information loss , error propagation and ignoring the int...
[ { "event_type": "ITT", "arguments": [ { "text": "relational triple extraction", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "relational", "triple", "extraction" ], "offsets": [ 0, 1, ...
[ "relational", "triple", "extraction", "is", "critical", "to", "understanding", "massive", "text", "corpora", "and", "constructing", "large", "-", "scale", "knowledge", "graph", ",", "which", "has", "attracted", "increasing", "research", "interest", ".", "however", ...
ACL
You Don’t Have Time to Read This: An Exploration of Document Reading Time Prediction
Predicting reading time has been a subject of much previous work, focusing on how different words affect human processing, measured by reading time. However, previous work has dealt with a limited number of participants as well as word level only predictions (i.e. predicting the time to read a single word). We seek to ...
ac48d3021f187a39adf127bd5e4f9f60
2,020
[ "predicting reading time has been a subject of much previous work , focusing on how different words affect human processing , measured by reading time .", "however , previous work has dealt with a limited number of participants as well as word level only predictions ( i . e . predicting the time to read a single ...
[ { "event_type": "ITT", "arguments": [ { "text": "predicting reading time", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "predicting", "reading", "time" ], "offsets": [ 0, 1, 2 ...
[ "predicting", "reading", "time", "has", "been", "a", "subject", "of", "much", "previous", "work", ",", "focusing", "on", "how", "different", "words", "affect", "human", "processing", ",", "measured", "by", "reading", "time", ".", "however", ",", "previous", ...
ACL
Accelerating BERT Inference for Sequence Labeling via Early-Exit
Both performance and efficiency are crucial factors for sequence labeling tasks in many real-world scenarios. Although the pre-trained models (PTMs) have significantly improved the performance of various sequence labeling tasks, their computational cost is expensive. To alleviate this problem, we extend the recent succ...
9bd0976ef2dd8ae235e8b4e6004581b4
2,021
[ "both performance and efficiency are crucial factors for sequence labeling tasks in many real - world scenarios .", "although the pre - trained models ( ptms ) have significantly improved the performance of various sequence labeling tasks , their computational cost is expensive .", "to alleviate this problem , ...
[ { "event_type": "ITT", "arguments": [ { "text": "sequence labeling", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "sequence", "labeling" ], "offsets": [ 8, 9 ] } ], "trigger": ...
[ "both", "performance", "and", "efficiency", "are", "crucial", "factors", "for", "sequence", "labeling", "tasks", "in", "many", "real", "-", "world", "scenarios", ".", "although", "the", "pre", "-", "trained", "models", "(", "ptms", ")", "have", "significantly"...
ACL
Improving Model Generalization: A Chinese Named Entity Recognition Case Study
Generalization is an important ability that helps to ensure that a machine learning model can perform well on unseen data. In this paper, we study the effect of data bias on model generalization, using Chinese Named Entity Recognition (NER) as a case study. Specifically, we analyzed five benchmarking datasets for Chine...
fd74f3ac9a92290f65759e3694020418
2,021
[ "generalization is an important ability that helps to ensure that a machine learning model can perform well on unseen data .", "in this paper , we study the effect of data bias on model generalization , using chinese named entity recognition ( ner ) as a case study .", "specifically , we analyzed five benchmark...
[ { "event_type": "ITT", "arguments": [ { "text": "generalization", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "generalization" ], "offsets": [ 0 ] } ], "trigger": { "text": "ability", ...
[ "generalization", "is", "an", "important", "ability", "that", "helps", "to", "ensure", "that", "a", "machine", "learning", "model", "can", "perform", "well", "on", "unseen", "data", ".", "in", "this", "paper", ",", "we", "study", "the", "effect", "of", "da...
ACL
Bird’s Eye: Probing for Linguistic Graph Structures with a Simple Information-Theoretic Approach
NLP has a rich history of representing our prior understanding of language in the form of graphs. Recent work on analyzing contextualized text representations has focused on hand-designed probe models to understand how and to what extent do these representations encode a particular linguistic phenomenon. However, due t...
d18d8cf2c74c62c1c96b341ebb63fe7e
2,021
[ "nlp has a rich history of representing our prior understanding of language in the form of graphs .", "recent work on analyzing contextualized text representations has focused on hand - designed probe models to understand how and to what extent do these representations encode a particular linguistic phenomenon ."...
[ { "event_type": "ITT", "arguments": [ { "text": "contextualized text representations", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "contextualized", "text", "representations" ], "offsets": [ 22, ...
[ "nlp", "has", "a", "rich", "history", "of", "representing", "our", "prior", "understanding", "of", "language", "in", "the", "form", "of", "graphs", ".", "recent", "work", "on", "analyzing", "contextualized", "text", "representations", "has", "focused", "on", "...
ACL
Context-aware Embedding for Targeted Aspect-based Sentiment Analysis
Attention-based neural models were employed to detect the different aspects and sentiment polarities of the same target in targeted aspect-based sentiment analysis (TABSA). However, existing methods do not specifically pre-train reasonable embeddings for targets and aspects in TABSA. This may result in targets or aspec...
4fc37d51832289fd00940c847d009a97
2,019
[ "attention - based neural models were employed to detect the different aspects and sentiment polarities of the same target in targeted aspect - based sentiment analysis ( tabsa ) .", "however , existing methods do not specifically pre - train reasonable embeddings for targets and aspects in tabsa .", "this may ...
[ { "event_type": "ITT", "arguments": [ { "text": "attention - based neural models", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "attention", "-", "based", "neural", "models" ], "offsets": ...
[ "attention", "-", "based", "neural", "models", "were", "employed", "to", "detect", "the", "different", "aspects", "and", "sentiment", "polarities", "of", "the", "same", "target", "in", "targeted", "aspect", "-", "based", "sentiment", "analysis", "(", "tabsa", ...
ACL
Coherent Comments Generation for Chinese Articles with a Graph-to-Sequence Model
Automatic article commenting is helpful in encouraging user engagement on online news platforms. However, the news documents are usually too long for models under traditional encoder-decoder frameworks, which often results in general and irrelevant comments. In this paper, we propose to generate comments with a graph-t...
a6d80475510fffadadcdc426c4f43282
2,019
[ "automatic article commenting is helpful in encouraging user engagement on online news platforms .", "however , the news documents are usually too long for models under traditional encoder - decoder frameworks , which often results in general and irrelevant comments .", "in this paper , we propose to generate c...
[ { "event_type": "ITT", "arguments": [ { "text": "automatic article commenting", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "automatic", "article", "commenting" ], "offsets": [ 0, 1, ...
[ "automatic", "article", "commenting", "is", "helpful", "in", "encouraging", "user", "engagement", "on", "online", "news", "platforms", ".", "however", ",", "the", "news", "documents", "are", "usually", "too", "long", "for", "models", "under", "traditional", "enc...
ACL
Neighborhood Matching Network for Entity Alignment
Structural heterogeneity between knowledge graphs is an outstanding challenge for entity alignment. This paper presents Neighborhood Matching Network (NMN), a novel entity alignment framework for tackling the structural heterogeneity challenge. NMN estimates the similarities between entities to capture both the topolog...
66b017a1d88850ef46742b3ecb84a45f
2,020
[ "structural heterogeneity between knowledge graphs is an outstanding challenge for entity alignment .", "this paper presents neighborhood matching network ( nmn ) , a novel entity alignment framework for tackling the structural heterogeneity challenge .", "nmn estimates the similarities between entities to capt...
[ { "event_type": "ITT", "arguments": [ { "text": "structural heterogeneity between knowledge graphs", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "structural", "heterogeneity", "between", "knowledge", "gr...
[ "structural", "heterogeneity", "between", "knowledge", "graphs", "is", "an", "outstanding", "challenge", "for", "entity", "alignment", ".", "this", "paper", "presents", "neighborhood", "matching", "network", "(", "nmn", ")", ",", "a", "novel", "entity", "alignment...
ACL
HAT: Hardware-Aware Transformers for Efficient Natural Language Processing
Transformers are ubiquitous in Natural Language Processing (NLP) tasks, but they are difficult to be deployed on hardware due to the intensive computation. To enable low-latency inference on resource-constrained hardware platforms, we propose to design Hardware-Aware Transformers (HAT) with neural architecture search. ...
0242ff24ca7bb1ccc0e092cab8189e35
2,020
[ "transformers are ubiquitous in natural language processing ( nlp ) tasks , but they are difficult to be deployed on hardware due to the intensive computation .", "to enable low - latency inference on resource - constrained hardware platforms , we propose to design hardware - aware transformers ( hat ) with neura...
[ { "event_type": "ITT", "arguments": [ { "text": "natural language processing", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "natural", "language", "processing" ], "offsets": [ 4, 5, ...
[ "transformers", "are", "ubiquitous", "in", "natural", "language", "processing", "(", "nlp", ")", "tasks", ",", "but", "they", "are", "difficult", "to", "be", "deployed", "on", "hardware", "due", "to", "the", "intensive", "computation", ".", "to", "enable", "...
ACL
Making Transformers Solve Compositional Tasks
Several studies have reported the inability of Transformer models to generalize compositionally, a key type of generalization in many NLP tasks such as semantic parsing. In this paper we explore the design space of Transformer models showing that the inductive biases given to the model by several design decisions signi...
102a436c728da5ecbafeb1fdeed4e984
2,022
[ "several studies have reported the inability of transformer models to generalize compositionally , a key type of generalization in many nlp tasks such as semantic parsing .", "in this paper we explore the design space of transformer models showing that the inductive biases given to the model by several design dec...
[ { "event_type": "RWF", "arguments": [ { "text": "inability of transformer models", "nugget_type": "WEA", "argument_type": "Fault", "tokens": [ "inability", "of", "transformer", "models" ], "offsets": [ 5,...
[ "several", "studies", "have", "reported", "the", "inability", "of", "transformer", "models", "to", "generalize", "compositionally", ",", "a", "key", "type", "of", "generalization", "in", "many", "nlp", "tasks", "such", "as", "semantic", "parsing", ".", "in", "...
ACL
Multi-Party Empathetic Dialogue Generation: A New Task for Dialog Systems
Empathetic dialogue assembles emotion understanding, feeling projection, and appropriate response generation. Existing work for empathetic dialogue generation concentrates on the two-party conversation scenario. Multi-party dialogues, however, are pervasive in reality. Furthermore, emotion and sensibility are typically...
b791e34651c1ff5cfed3953cb41dda32
2,022
[ "empathetic dialogue assembles emotion understanding , feeling projection , and appropriate response generation .", "existing work for empathetic dialogue generation concentrates on the two - party conversation scenario .", "multi - party dialogues , however , are pervasive in reality .", "furthermore , emoti...
[ { "event_type": "ITT", "arguments": [ { "text": "empathetic dialogue", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "empathetic", "dialogue" ], "offsets": [ 0, 1 ] } ], "trigge...
[ "empathetic", "dialogue", "assembles", "emotion", "understanding", ",", "feeling", "projection", ",", "and", "appropriate", "response", "generation", ".", "existing", "work", "for", "empathetic", "dialogue", "generation", "concentrates", "on", "the", "two", "-", "pa...
ACL
Learning to Imagine: Integrating Counterfactual Thinking in Neural Discrete Reasoning
Neural discrete reasoning (NDR) has shown remarkable progress in combining deep models with discrete reasoning. However, we find that existing NDR solution suffers from large performance drop on hypothetical questions, e.g. “what the annualized rate of return would be if the revenue in 2020 was doubled”. The key to hyp...
9f92fd995bcf732ced30de3fb8164923
2,022
[ "neural discrete reasoning ( ndr ) has shown remarkable progress in combining deep models with discrete reasoning .", "however , we find that existing ndr solution suffers from large performance drop on hypothetical questions , e . g . “ what the annualized rate of return would be if the revenue in 2020 was doubl...
[ { "event_type": "ITT", "arguments": [ { "text": "neural discrete reasoning", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "neural", "discrete", "reasoning" ], "offsets": [ 0, 1, ...
[ "neural", "discrete", "reasoning", "(", "ndr", ")", "has", "shown", "remarkable", "progress", "in", "combining", "deep", "models", "with", "discrete", "reasoning", ".", "however", ",", "we", "find", "that", "existing", "ndr", "solution", "suffers", "from", "la...
ACL
Do You Know That Florence Is Packed with Visitors? Evaluating State-of-the-art Models of Speaker Commitment
When a speaker, Mary, asks “Do you know that Florence is packed with visitors?”, we take her to believe that Florence is packed with visitors, but not if she asks “Do you think that Florence is packed with visitors?”. Inferring speaker commitment (aka event factuality) is crucial for information extraction and question...
067896d775c9eb9c0d3c3372aeda9e33
2,019
[ "when a speaker , mary , asks “ do you know that florence is packed with visitors ? ” , we take her to believe that florence is packed with visitors , but not if she asks “ do you think that florence is packed with visitors ? ” .", "inferring speaker commitment ( aka event factuality ) is crucial for information ...
[ { "event_type": "ITT", "arguments": [ { "text": "inferring speaker commitment", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "inferring", "speaker", "commitment" ], "offsets": [ 50, 51, ...
[ "when", "a", "speaker", ",", "mary", ",", "asks", "“", "do", "you", "know", "that", "florence", "is", "packed", "with", "visitors", "?", "”", ",", "we", "take", "her", "to", "believe", "that", "florence", "is", "packed", "with", "visitors", ",", "but",...
ACL
Identifying Moments of Change from Longitudinal User Text
Identifying changes in individuals’ behaviour and mood, as observed via content shared on online platforms, is increasingly gaining importance. Most research to-date on this topic focuses on either: (a) identifying individuals at risk or with a certain mental health condition given a batch of posts or (b) providing equ...
eadc5a6cc97199aa115a6ba6379830a0
2,022
[ "identifying changes in individuals ’ behaviour and mood , as observed via content shared on online platforms , is increasingly gaining importance .", "most research to - date on this topic focuses on either : ( a ) identifying individuals at risk or with a certain mental health condition given a batch of posts o...
[ { "event_type": "ITT", "arguments": [ { "text": "identifying changes in individuals ’ behaviour and mood", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "identifying", "changes", "in", "individuals", "’", ...
[ "identifying", "changes", "in", "individuals", "’", "behaviour", "and", "mood", ",", "as", "observed", "via", "content", "shared", "on", "online", "platforms", ",", "is", "increasingly", "gaining", "importance", ".", "most", "research", "to", "-", "date", "on"...
ACL
Structural Guidance for Transformer Language Models
Transformer-based language models pre-trained on large amounts of text data have proven remarkably successful in learning generic transferable linguistic representations. Here we study whether structural guidance leads to more human-like systematic linguistic generalization in Transformer language models without resort...
0114ab91a7434d6cfd21ff863a6bb40a
2,021
[ "transformer - based language models pre - trained on large amounts of text data have proven remarkably successful in learning generic transferable linguistic representations .", "here we study whether structural guidance leads to more human - like systematic linguistic generalization in transformer language mode...
[ { "event_type": "ITT", "arguments": [ { "text": "transformer - based language models", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "transformer", "-", "based", "language", "models" ], "of...
[ "transformer", "-", "based", "language", "models", "pre", "-", "trained", "on", "large", "amounts", "of", "text", "data", "have", "proven", "remarkably", "successful", "in", "learning", "generic", "transferable", "linguistic", "representations", ".", "here", "we",...
ACL
The Moral Debater: A Study on the Computational Generation of Morally Framed Arguments
An audience’s prior beliefs and morals are strong indicators of how likely they will be affected by a given argument. Utilizing such knowledge can help focus on shared values to bring disagreeing parties towards agreement. In argumentation technology, however, this is barely exploited so far. This paper studies the fea...
a722a6ae48c3e448c2f9f62b80a7fee3
2,022
[ "an audience ’ s prior beliefs and morals are strong indicators of how likely they will be affected by a given argument .", "utilizing such knowledge can help focus on shared values to bring disagreeing parties towards agreement .", "in argumentation technology , however , this is barely exploited so far .", ...
[ { "event_type": "WKS", "arguments": [ { "text": "feasibility of automatically generating morally framed arguments", "nugget_type": "TAK", "argument_type": "Content", "tokens": [ "feasibility", "of", "automatically", "generating", ...
[ "an", "audience", "’", "s", "prior", "beliefs", "and", "morals", "are", "strong", "indicators", "of", "how", "likely", "they", "will", "be", "affected", "by", "a", "given", "argument", ".", "utilizing", "such", "knowledge", "can", "help", "focus", "on", "s...
ACL
Learning Compressed Sentence Representations for On-Device Text Processing
Vector representations of sentences, trained on massive text corpora, are widely used as generic sentence embeddings across a variety of NLP problems. The learned representations are generally assumed to be continuous and real-valued, giving rise to a large memory footprint and slow retrieval speed, which hinders their...
31f243a45d294fc0f6b4a1b5e365a74c
2,019
[ "vector representations of sentences , trained on massive text corpora , are widely used as generic sentence embeddings across a variety of nlp problems .", "the learned representations are generally assumed to be continuous and real - valued , giving rise to a large memory footprint and slow retrieval speed , wh...
[ { "event_type": "ITT", "arguments": [ { "text": "vector representations of sentences", "nugget_type": "FEA", "argument_type": "Target", "tokens": [ "vector", "representations", "of", "sentences" ], "offsets": [ ...
[ "vector", "representations", "of", "sentences", ",", "trained", "on", "massive", "text", "corpora", ",", "are", "widely", "used", "as", "generic", "sentence", "embeddings", "across", "a", "variety", "of", "nlp", "problems", ".", "the", "learned", "representation...
ACL
Structure-Level Knowledge Distillation For Multilingual Sequence Labeling
Multilingual sequence labeling is a task of predicting label sequences using a single unified model for multiple languages. Compared with relying on multiple monolingual models, using a multilingual model has the benefit of a smaller model size, easier in online serving, and generalizability to low-resource languages. ...
e76f657dc7d9e794204da0499cff77e9
2,020
[ "multilingual sequence labeling is a task of predicting label sequences using a single unified model for multiple languages .", "compared with relying on multiple monolingual models , using a multilingual model has the benefit of a smaller model size , easier in online serving , and generalizability to low - reso...
[ { "event_type": "ITT", "arguments": [ { "text": "multilingual sequence labeling", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "multilingual", "sequence", "labeling" ], "offsets": [ 0, 1, ...
[ "multilingual", "sequence", "labeling", "is", "a", "task", "of", "predicting", "label", "sequences", "using", "a", "single", "unified", "model", "for", "multiple", "languages", ".", "compared", "with", "relying", "on", "multiple", "monolingual", "models", ",", "...
ACL
ClusterFormer: Neural Clustering Attention for Efficient and Effective Transformer
Recently, a lot of research has been carried out to improve the efficiency of Transformer. Among them, the sparse pattern-based method is an important branch of efficient Transformers. However, some existing sparse methods usually use fixed patterns to select words, without considering similarities between words. Other...
4a206aa908fa42c6f439eb7d309dd318
2,022
[ "recently , a lot of research has been carried out to improve the efficiency of transformer .", "among them , the sparse pattern - based method is an important branch of efficient transformers .", "however , some existing sparse methods usually use fixed patterns to select words , without considering similariti...
[ { "event_type": "ITT", "arguments": [ { "text": "efficient transformers", "nugget_type": "MOD", "argument_type": "Target", "tokens": [ "efficient", "transformers" ], "offsets": [ 31, 32 ] } ], ...
[ "recently", ",", "a", "lot", "of", "research", "has", "been", "carried", "out", "to", "improve", "the", "efficiency", "of", "transformer", ".", "among", "them", ",", "the", "sparse", "pattern", "-", "based", "method", "is", "an", "important", "branch", "of...
ACL
Predicting Human Activities from User-Generated Content
The activities we do are linked to our interests, personality, political preferences, and decisions we make about the future. In this paper, we explore the task of predicting human activities from user-generated content. We collect a dataset containing instances of social media users writing about a range of everyday a...
a392b7e0b16c0f5158bf4afdc0faa50a
2,019
[ "the activities we do are linked to our interests , personality , political preferences , and decisions we make about the future .", "in this paper , we explore the task of predicting human activities from user - generated content .", "we collect a dataset containing instances of social media users writing abou...
[ { "event_type": "WKS", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Researcher", "tokens": [ "we" ], "offsets": [ 27 ] }, { "text": "task of predicting human activities", "n...
[ "the", "activities", "we", "do", "are", "linked", "to", "our", "interests", ",", "personality", ",", "political", "preferences", ",", "and", "decisions", "we", "make", "about", "the", "future", ".", "in", "this", "paper", ",", "we", "explore", "the", "task...
ACL
Taxonomy Construction of Unseen Domains via Graph-based Cross-Domain Knowledge Transfer
Extracting lexico-semantic relations as graph-structured taxonomies, also known as taxonomy construction, has been beneficial in a variety of NLP applications. Recently Graph Neural Network (GNN) has shown to be powerful in successfully tackling many tasks. However, there has been no attempt to exploit GNN to create ta...
3887ebeeb34ef4caedac86d99e48fa66
2,020
[ "extracting lexico - semantic relations as graph - structured taxonomies , also known as taxonomy construction , has been beneficial in a variety of nlp applications .", "recently graph neural network ( gnn ) has shown to be powerful in successfully tackling many tasks .", "however , there has been no attempt t...
[ { "event_type": "ITT", "arguments": [ { "text": "graph neural network", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "graph", "neural", "network" ], "offsets": [ 28, 29, 30 ...
[ "extracting", "lexico", "-", "semantic", "relations", "as", "graph", "-", "structured", "taxonomies", ",", "also", "known", "as", "taxonomy", "construction", ",", "has", "been", "beneficial", "in", "a", "variety", "of", "nlp", "applications", ".", "recently", ...
ACL
Improving Visual Question Answering by Referring to Generated Paragraph Captions
Paragraph-style image captions describe diverse aspects of an image as opposed to the more common single-sentence captions that only provide an abstract description of the image. These paragraph captions can hence contain substantial information of the image for tasks such as visual question answering. Moreover, this t...
2dbea8150c688e3748a186cc0c285496
2,019
[ "paragraph - style image captions describe diverse aspects of an image as opposed to the more common single - sentence captions that only provide an abstract description of the image .", "these paragraph captions can hence contain substantial information of the image for tasks such as visual question answering ."...
[ { "event_type": "ITT", "arguments": [ { "text": "paragraph - style image captions", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "paragraph", "-", "style", "image", "captions" ], "offsets"...
[ "paragraph", "-", "style", "image", "captions", "describe", "diverse", "aspects", "of", "an", "image", "as", "opposed", "to", "the", "more", "common", "single", "-", "sentence", "captions", "that", "only", "provide", "an", "abstract", "description", "of", "the...
ACL
Discrete Optimization for Unsupervised Sentence Summarization with Word-Level Extraction
Automatic sentence summarization produces a shorter version of a sentence, while preserving its most important information. A good summary is characterized by language fluency and high information overlap with the source sentence. We model these two aspects in an unsupervised objective function, consisting of language ...
e76860b37b333c670551e38d7f29b1f8
2,020
[ "automatic sentence summarization produces a shorter version of a sentence , while preserving its most important information .", "a good summary is characterized by language fluency and high information overlap with the source sentence .", "we model these two aspects in an unsupervised objective function , cons...
[ { "event_type": "ITT", "arguments": [ { "text": "automatic sentence summarization", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "automatic", "sentence", "summarization" ], "offsets": [ 0, ...
[ "automatic", "sentence", "summarization", "produces", "a", "shorter", "version", "of", "a", "sentence", ",", "while", "preserving", "its", "most", "important", "information", ".", "a", "good", "summary", "is", "characterized", "by", "language", "fluency", "and", ...
ACL
An Empirical Study of Span Representations in Argumentation Structure Parsing
For several natural language processing (NLP) tasks, span representation design is attracting considerable attention as a promising new technique; a common basis for an effective design has been established. With such basis, exploring task-dependent extensions for argumentation structure parsing (ASP) becomes an intere...
f24c802f613874cac5f5f8b9b63f588d
2,019
[ "for several natural language processing ( nlp ) tasks , span representation design is attracting considerable attention as a promising new technique ; a common basis for an effective design has been established .", "with such basis , exploring task - dependent extensions for argumentation structure parsing ( asp...
[ { "event_type": "ITT", "arguments": [ { "text": "argumentation structure parsing", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "argumentation", "structure", "parsing" ], "offsets": [ 44, ...
[ "for", "several", "natural", "language", "processing", "(", "nlp", ")", "tasks", ",", "span", "representation", "design", "is", "attracting", "considerable", "attention", "as", "a", "promising", "new", "technique", ";", "a", "common", "basis", "for", "an", "ef...
ACL
MISC: A Mixed Strategy-Aware Model integrating COMET for Emotional Support Conversation
Applying existing methods to emotional support conversation—which provides valuable assistance to people who are in need—has two major limitations: (a) they generally employ a conversation-level emotion label, which is too coarse-grained to capture user’s instant mental state; (b) most of them focus on expressing empat...
d48c01bffd85aedc8b34bdae5e42ad8a
2,022
[ "applying existing methods to emotional support conversation — which provides valuable assistance to people who are in need — has two major limitations : ( a ) they generally employ a conversation - level emotion label , which is too coarse - grained to capture user ’ s instant mental state ; ( b ) most of them foc...
[ { "event_type": "ITT", "arguments": [ { "text": "emotional support conversation", "nugget_type": "FEA", "argument_type": "Target", "tokens": [ "emotional", "support", "conversation" ], "offsets": [ 4, 5, ...
[ "applying", "existing", "methods", "to", "emotional", "support", "conversation", "—", "which", "provides", "valuable", "assistance", "to", "people", "who", "are", "in", "need", "—", "has", "two", "major", "limitations", ":", "(", "a", ")", "they", "generally",...
ACL
Primum Non Nocere: Before working with Indigenous data, the ACL must confront ongoing colonialism
In this paper, we challenge the ACL community to reckon with historical and ongoing colonialism by adopting a set of ethical obligations and best practices drawn from the Indigenous studies literature. While the vast majority of NLP research focuses on a very small number of very high resource languages (English, Chine...
180761257e17159faf2a530e2d537f09
2,022
[ "in this paper , we challenge the acl community to reckon with historical and ongoing colonialism by adopting a set of ethical obligations and best practices drawn from the indigenous studies literature .", "while the vast majority of nlp research focuses on a very small number of very high resource languages ( e...
[ { "event_type": "WKS", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Researcher", "tokens": [ "we" ], "offsets": [ 4 ] }, { "text": "acl community", "nugget_type": "TAK", ...
[ "in", "this", "paper", ",", "we", "challenge", "the", "acl", "community", "to", "reckon", "with", "historical", "and", "ongoing", "colonialism", "by", "adopting", "a", "set", "of", "ethical", "obligations", "and", "best", "practices", "drawn", "from", "the", ...
ACL
Language-aware Interlingua for Multilingual Neural Machine Translation
Multilingual neural machine translation (NMT) has led to impressive accuracy improvements in low-resource scenarios by sharing common linguistic information across languages. However, the traditional multilingual model fails to capture the diversity and specificity of different languages, resulting in inferior performa...
b703e96f2bfd4ad4dacaf62b299e06b7
2,020
[ "multilingual neural machine translation ( nmt ) has led to impressive accuracy improvements in low - resource scenarios by sharing common linguistic information across languages .", "however , the traditional multilingual model fails to capture the diversity and specificity of different languages , resulting in ...
[ { "event_type": "ITT", "arguments": [ { "text": "multilingual neural machine translation", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "multilingual", "neural", "machine", "translation" ], "offsets...
[ "multilingual", "neural", "machine", "translation", "(", "nmt", ")", "has", "led", "to", "impressive", "accuracy", "improvements", "in", "low", "-", "resource", "scenarios", "by", "sharing", "common", "linguistic", "information", "across", "languages", ".", "howev...
ACL
GraphRel: Modeling Text as Relational Graphs for Joint Entity and Relation Extraction
In this paper, we present GraphRel, an end-to-end relation extraction model which uses graph convolutional networks (GCNs) to jointly learn named entities and relations. In contrast to previous baselines, we consider the interaction between named entities and relations via a 2nd-phase relation-weighted GCN to better ex...
26d53382a8fbb1565a7ec063857055a1
2,019
[ "in this paper , we present graphrel , an end - to - end relation extraction model which uses graph convolutional networks ( gcns ) to jointly learn named entities and relations .", "in contrast to previous baselines , we consider the interaction between named entities and relations via a 2nd - phase relation - w...
[ { "event_type": "FAC", "arguments": [ { "text": "end - to - end relation extraction model", "nugget_type": "APP", "argument_type": "Subject", "tokens": [ "end", "-", "to", "-", "end", "relation", "ext...
[ "in", "this", "paper", ",", "we", "present", "graphrel", ",", "an", "end", "-", "to", "-", "end", "relation", "extraction", "model", "which", "uses", "graph", "convolutional", "networks", "(", "gcns", ")", "to", "jointly", "learn", "named", "entities", "an...
ACL
Successfully Applying the Stabilized Lottery Ticket Hypothesis to the Transformer Architecture
Sparse models require less memory for storage and enable a faster inference by reducing the necessary number of FLOPs. This is relevant both for time-critical and on-device computations using neural networks. The stabilized lottery ticket hypothesis states that networks can be pruned after none or few training iteratio...
12d2dab6122d4d85e2e9bcf90c4ecccb
2,020
[ "sparse models require less memory for storage and enable a faster inference by reducing the necessary number of flops .", "this is relevant both for time - critical and on - device computations using neural networks .", "the stabilized lottery ticket hypothesis states that networks can be pruned after none or ...
[ { "event_type": "ITT", "arguments": [ { "text": "sparse models", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "sparse", "models" ], "offsets": [ 0, 1 ] } ], "trigger": { ...
[ "sparse", "models", "require", "less", "memory", "for", "storage", "and", "enable", "a", "faster", "inference", "by", "reducing", "the", "necessary", "number", "of", "flops", ".", "this", "is", "relevant", "both", "for", "time", "-", "critical", "and", "on",...
ACL
Interconnected Question Generation with Coreference Alignment and Conversation Flow Modeling
We study the problem of generating interconnected questions in question-answering style conversations. Compared with previous works which generate questions based on a single sentence (or paragraph), this setting is different in two major aspects: (1) Questions are highly conversational. Almost half of them refer back ...
3c2291327509c186b9b6c2ecc280db9d
2,019
[ "we study the problem of generating interconnected questions in question - answering style conversations .", "compared with previous works which generate questions based on a single sentence ( or paragraph ) , this setting is different in two major aspects : ( 1 ) questions are highly conversational .", "almost...
[ { "event_type": "ITT", "arguments": [ { "text": "generating interconnected questions", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "generating", "interconnected", "questions" ], "offsets": [ 5, ...
[ "we", "study", "the", "problem", "of", "generating", "interconnected", "questions", "in", "question", "-", "answering", "style", "conversations", ".", "compared", "with", "previous", "works", "which", "generate", "questions", "based", "on", "a", "single", "sentenc...
ACL
How Does Selective Mechanism Improve Self-Attention Networks?
Self-attention networks (SANs) with selective mechanism has produced substantial improvements in various NLP tasks by concentrating on a subset of input words. However, the underlying reasons for their strong performance have not been well explained. In this paper, we bridge the gap by assessing the strengths of select...
8894399dd6235640028acd5310cb3859
2,020
[ "self - attention networks ( sans ) with selective mechanism has produced substantial improvements in various nlp tasks by concentrating on a subset of input words .", "however , the underlying reasons for their strong performance have not been well explained .", "in this paper , we bridge the gap by assessing ...
[ { "event_type": "ITT", "arguments": [ { "text": "self - attention networks", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "self", "-", "attention", "networks" ], "offsets": [ 0, ...
[ "self", "-", "attention", "networks", "(", "sans", ")", "with", "selective", "mechanism", "has", "produced", "substantial", "improvements", "in", "various", "nlp", "tasks", "by", "concentrating", "on", "a", "subset", "of", "input", "words", ".", "however", ","...
ACL
SafetyKit: First Aid for Measuring Safety in Open-domain Conversational Systems
The social impact of natural language processing and its applications has received increasing attention. In this position paper, we focus on the problem of safety for end-to-end conversational AI. We survey the problem landscape therein, introducing a taxonomy of three observed phenomena: the Instigator, Yea-Sayer, and...
b0c426a81a9a5a62ea1d26a0457bbe96
2,022
[ "the social impact of natural language processing and its applications has received increasing attention .", "in this position paper , we focus on the problem of safety for end - to - end conversational ai .", "we survey the problem landscape therein , introducing a taxonomy of three observed phenomena : the in...
[ { "event_type": "ITT", "arguments": [ { "text": "end - to - end conversational ai", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "end", "-", "to", "-", "end", "conversational", "ai" ...
[ "the", "social", "impact", "of", "natural", "language", "processing", "and", "its", "applications", "has", "received", "increasing", "attention", ".", "in", "this", "position", "paper", ",", "we", "focus", "on", "the", "problem", "of", "safety", "for", "end", ...
ACL
RepSum: Unsupervised Dialogue Summarization based on Replacement Strategy
In the field of dialogue summarization, due to the lack of training data, it is often difficult for supervised summary generation methods to learn vital information from dialogue context with limited data. Several attempts on unsupervised summarization for text by leveraging semantic information solely or auto-encoder ...
3636df92a05af6d5b1d5fab28173c591
2,021
[ "in the field of dialogue summarization , due to the lack of training data , it is often difficult for supervised summary generation methods to learn vital information from dialogue context with limited data .", "several attempts on unsupervised summarization for text by leveraging semantic information solely or ...
[ { "event_type": "ITT", "arguments": [ { "text": "dialogue summarization", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "dialogue", "summarization" ], "offsets": [ 4, 5 ] } ], "...
[ "in", "the", "field", "of", "dialogue", "summarization", ",", "due", "to", "the", "lack", "of", "training", "data", ",", "it", "is", "often", "difficult", "for", "supervised", "summary", "generation", "methods", "to", "learn", "vital", "information", "from", ...
ACL
Graph Pre-training for AMR Parsing and Generation
Abstract meaning representation (AMR) highlights the core semantic information of text in a graph structure.Recently, pre-trained language models (PLMs) have advanced tasks of AMR parsing and AMR-to-text generation, respectively.However, PLMs are typically pre-trained on textual data, thus are sub-optimal for modeling ...
16928d6e37ed36ab27e9d643c6d4d587
2,022
[ "abstract meaning representation ( amr ) highlights the core semantic information of text in a graph structure .", "recently , pre - trained language models ( plms ) have advanced tasks of amr parsing and amr - to - text generation , respectively .", "however , plms are typically pre - trained on textual data ,...
[ { "event_type": "ITT", "arguments": [ { "text": "amr", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "amr" ], "offsets": [ 148 ] } ], "trigger": { "text": "highlights", "tokens": [ ...
[ "abstract", "meaning", "representation", "(", "amr", ")", "highlights", "the", "core", "semantic", "information", "of", "text", "in", "a", "graph", "structure", ".", "recently", ",", "pre", "-", "trained", "language", "models", "(", "plms", ")", "have", "adv...
ACL
Shape of Synth to Come: Why We Should Use Synthetic Data for English Surface Realization
The Surface Realization Shared Tasks of 2018 and 2019 were Natural Language Generation shared tasks with the goal of exploring approaches to surface realization from Universal-Dependency-like trees to surface strings for several languages. In the 2018 shared task there was very little difference in the absolute perform...
e3c2eb419c6f6e145640690046f77c70
2,020
[ "the surface realization shared tasks of 2018 and 2019 were natural language generation shared tasks with the goal of exploring approaches to surface realization from universal - dependency - like trees to surface strings for several languages .", "in the 2018 shared task there was very little difference in the a...
[ { "event_type": "ITT", "arguments": [ { "text": "natural language generation shared tasks", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "natural", "language", "generation", "shared", "tasks" ], ...
[ "the", "surface", "realization", "shared", "tasks", "of", "2018", "and", "2019", "were", "natural", "language", "generation", "shared", "tasks", "with", "the", "goal", "of", "exploring", "approaches", "to", "surface", "realization", "from", "universal", "-", "de...
ACL
A DQN-based Approach to Finding Precise Evidences for Fact Verification
Computing precise evidences, namely minimal sets of sentences that support or refute a given claim, rather than larger evidences is crucial in fact verification (FV), since larger evidences may contain conflicting pieces some of which support the claim while the other refute, thereby misleading FV. Despite being import...
7755c955a9031c280cc4eab1a1fd6f8b
2,021
[ "computing precise evidences , namely minimal sets of sentences that support or refute a given claim , rather than larger evidences is crucial in fact verification ( fv ) , since larger evidences may contain conflicting pieces some of which support the claim while the other refute , thereby misleading fv .", "des...
[ { "event_type": "ITT", "arguments": [ { "text": "computing precise evidences", "nugget_type": "FEA", "argument_type": "Target", "tokens": [ "computing", "precise", "evidences" ], "offsets": [ 0, 1, ...
[ "computing", "precise", "evidences", ",", "namely", "minimal", "sets", "of", "sentences", "that", "support", "or", "refute", "a", "given", "claim", ",", "rather", "than", "larger", "evidences", "is", "crucial", "in", "fact", "verification", "(", "fv", ")", "...
ACL
Exact yet Efficient Graph Parsing, Bi-directional Locality and the Constructivist Hypothesis
A key problem in processing graph-based meaning representations is graph parsing, i.e. computing all possible derivations of a given graph according to a (competence) grammar. We demonstrate, for the first time, that exact graph parsing can be efficient for large graphs and with large Hyperedge Replacement Grammars (HR...
79753fec5b40bcba7e4f69f42bc3f1be
2,020
[ "a key problem in processing graph - based meaning representations is graph parsing , i . e . computing all possible derivations of a given graph according to a ( competence ) grammar .", "we demonstrate , for the first time , that exact graph parsing can be efficient for large graphs and with large hyperedge rep...
[ { "event_type": "ITT", "arguments": [ { "text": "graph - based meaning representations", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "graph", "-", "based", "meaning", "representations" ], ...
[ "a", "key", "problem", "in", "processing", "graph", "-", "based", "meaning", "representations", "is", "graph", "parsing", ",", "i", ".", "e", ".", "computing", "all", "possible", "derivations", "of", "a", "given", "graph", "according", "to", "a", "(", "com...
ACL
Which side are you on? Insider-Outsider classification in conspiracy-theoretic social media
Social media is a breeding ground for threat narratives and related conspiracy theories. In these, an outside group threatens the integrity of an inside group, leading to the emergence of sharply defined group identities: Insiders – agents with whom the authors identify and Outsiders – agents who threaten the insiders....
87b0f9e8655fd5853caf2e4178ce21ab
2,022
[ "social media is a breeding ground for threat narratives and related conspiracy theories .", "in these , an outside group threatens the integrity of an inside group , leading to the emergence of sharply defined group identities : insiders – agents with whom the authors identify and outsiders – agents who threaten...
[ { "event_type": "ITT", "arguments": [ { "text": "social media", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "social", "media" ], "offsets": [ 0, 1 ] } ], "trigger": { "t...
[ "social", "media", "is", "a", "breeding", "ground", "for", "threat", "narratives", "and", "related", "conspiracy", "theories", ".", "in", "these", ",", "an", "outside", "group", "threatens", "the", "integrity", "of", "an", "inside", "group", ",", "leading", ...
ACL
Stock Embeddings Acquired from News Articles and Price History, and an Application to Portfolio Optimization
Previous works that integrated news articles to better process stock prices used a variety of neural networks to predict price movements. The textual and price information were both encoded in the neural network, and it is therefore difficult to apply this approach in situations other than the original framework of the...
615e9285fa36be78d41854eb6783a62d
2,020
[ "previous works that integrated news articles to better process stock prices used a variety of neural networks to predict price movements .", "the textual and price information were both encoded in the neural network , and it is therefore difficult to apply this approach in situations other than the original fram...
[ { "event_type": "RWS", "arguments": [ { "text": "news articles", "nugget_type": "FEA", "argument_type": "TriedComponent", "tokens": [ "news", "articles" ], "offsets": [ 4, 5 ] }, { "text...
[ "previous", "works", "that", "integrated", "news", "articles", "to", "better", "process", "stock", "prices", "used", "a", "variety", "of", "neural", "networks", "to", "predict", "price", "movements", ".", "the", "textual", "and", "price", "information", "were", ...
ACL
Towards Scalable and Reliable Capsule Networks for Challenging NLP Applications
Obstacles hindering the development of capsule networks for challenging NLP applications include poor scalability to large output spaces and less reliable routing processes. In this paper, we introduce: (i) an agreement score to evaluate the performance of routing processes at instance-level; (ii) an adaptive optimizer...
778648f5ad26c4a5bcae13cb0595045d
2,019
[ "obstacles hindering the development of capsule networks for challenging nlp applications include poor scalability to large output spaces and less reliable routing processes .", "in this paper , we introduce :", "( i ) an agreement score to evaluate the performance of routing processes at instance - level ; ( i...
[ { "event_type": "RWF", "arguments": [ { "text": "capsule networks", "nugget_type": "APP", "argument_type": "Concern", "tokens": [ "capsule", "networks" ], "offsets": [ 5, 6 ] }, { "text"...
[ "obstacles", "hindering", "the", "development", "of", "capsule", "networks", "for", "challenging", "nlp", "applications", "include", "poor", "scalability", "to", "large", "output", "spaces", "and", "less", "reliable", "routing", "processes", ".", "in", "this", "pa...
ACL
An Effective Transition-based Model for Discontinuous NER
Unlike widely used Named Entity Recognition (NER) data sets in generic domains, biomedical NER data sets often contain mentions consisting of discontinuous spans. Conventional sequence tagging techniques encode Markov assumptions that are efficient but preclude recovery of these mentions. We propose a simple, effective...
e333e855cb86663625d3b19700f24863
2,020
[ "unlike widely used named entity recognition ( ner ) data sets in generic domains , biomedical ner data sets often contain mentions consisting of discontinuous spans .", "conventional sequence tagging techniques encode markov assumptions that are efficient but preclude recovery of these mentions .", "we propose...
[ { "event_type": "ITT", "arguments": [ { "text": "biomedical named entity recognition data sets", "nugget_type": "DST", "argument_type": "Target", "tokens": [ "biomedical", "named", "entity", "recognition", "data", ...
[ "unlike", "widely", "used", "named", "entity", "recognition", "(", "ner", ")", "data", "sets", "in", "generic", "domains", ",", "biomedical", "ner", "data", "sets", "often", "contain", "mentions", "consisting", "of", "discontinuous", "spans", ".", "conventional"...
ACL
From Machine Translation to Code-Switching: Generating High-Quality Code-Switched Text
Generating code-switched text is a problem of growing interest, especially given the scarcity of corpora containing large volumes of real code-switched text. In this work, we adapt a state-of-the-art neural machine translation model to generate Hindi-English code-switched sentences starting from monolingual Hindi sente...
ed676bd77837a0e0ab1bac581463f6e6
2,021
[ "generating code - switched text is a problem of growing interest , especially given the scarcity of corpora containing large volumes of real code - switched text .", "in this work , we adapt a state - of - the - art neural machine translation model to generate hindi - english code - switched sentences starting f...
[ { "event_type": "ITT", "arguments": [ { "text": "code - switched text", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "code", "-", "switched", "text" ], "offsets": [ 1, 2, ...
[ "generating", "code", "-", "switched", "text", "is", "a", "problem", "of", "growing", "interest", ",", "especially", "given", "the", "scarcity", "of", "corpora", "containing", "large", "volumes", "of", "real", "code", "-", "switched", "text", ".", "in", "thi...
ACL
DTCA: Decision Tree-based Co-Attention Networks for Explainable Claim Verification
Recently, many methods discover effective evidence from reliable sources by appropriate neural networks for explainable claim verification, which has been widely recognized. However, in these methods, the discovery process of evidence is nontransparent and unexplained. Simultaneously, the discovered evidence is aimed a...
dc9a3ef5b54266536b8408df2925919e
2,020
[ "recently , many methods discover effective evidence from reliable sources by appropriate neural networks for explainable claim verification , which has been widely recognized .", "however , in these methods , the discovery process of evidence is nontransparent and unexplained .", "simultaneously , the discover...
[ { "event_type": "RWF", "arguments": [ { "text": "unexplained", "nugget_type": "WEA", "argument_type": "Fault", "tokens": [ "unexplained" ], "offsets": [ 39 ] }, { "text": "nontransparent", "nugget_t...
[ "recently", ",", "many", "methods", "discover", "effective", "evidence", "from", "reliable", "sources", "by", "appropriate", "neural", "networks", "for", "explainable", "claim", "verification", ",", "which", "has", "been", "widely", "recognized", ".", "however", "...
ACL
Introducing Orthogonal Constraint in Structural Probes
With the recent success of pre-trained models in NLP, a significant focus was put on interpreting their representations. One of the most prominent approaches is structural probing (Hewitt and Manning, 2019), where a linear projection of word embeddings is performed in order to approximate the topology of dependency str...
e820e1e7c761e3f6ad7fe2e177eefd54
2,021
[ "with the recent success of pre - trained models in nlp , a significant focus was put on interpreting their representations .", "one of the most prominent approaches is structural probing ( hewitt and manning , 2019 ) , where a linear projection of word embeddings is performed in order to approximate the topology...
[ { "event_type": "ITT", "arguments": [ { "text": "pre - trained models", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "pre", "-", "trained", "models" ], "offsets": [ 5, 6, ...
[ "with", "the", "recent", "success", "of", "pre", "-", "trained", "models", "in", "nlp", ",", "a", "significant", "focus", "was", "put", "on", "interpreting", "their", "representations", ".", "one", "of", "the", "most", "prominent", "approaches", "is", "struc...