DataHammer/SciEvents · Datasets at Hugging Face

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	A Lightweight Recurrent Network for Sequence Modeling	Recurrent networks have achieved great success on various sequential tasks with the assistance of complex recurrent units, but suffer from severe computational inefficiency due to weak parallelization. One direction to alleviate this issue is to shift heavy computations outside the recurrence. In this paper, we propose...	e7d6faf5905e216a5d25cc8404b063c9	2,019	[ "recurrent networks have achieved great success on various sequential tasks with the assistance of complex recurrent units , but suffer from severe computational inefficiency due to weak parallelization .", "one direction to alleviate this issue is to shift heavy computations outside the recurrence .", "in this...	[ { "event_type": "ITT", "arguments": [ { "text": "sequential tasks", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "sequential", "tasks" ], "offsets": [ 8, 9 ] } ], "trigger": { ...	[ "recurrent", "networks", "have", "achieved", "great", "success", "on", "various", "sequential", "tasks", "with", "the", "assistance", "of", "complex", "recurrent", "units", ",", "but", "suffer", "from", "severe", "computational", "inefficiency", "due", "to", "weak...
ACL	Contrastive Self-Supervised Learning for Commonsense Reasoning	We propose a self-supervised method to solve Pronoun Disambiguation and Winograd Schema Challenge problems. Our approach exploits the characteristic structure of training corpora related to so-called “trigger” words, which are responsible for flipping the answer in pronoun disambiguation. We achieve such commonsense re...	67aee4c4326695393662a0a4c6051845	2,020	[ "we propose a self - supervised method to solve pronoun disambiguation and winograd schema challenge problems .", "our approach exploits the characteristic structure of training corpora related to so - called “ trigger ” words , which are responsible for flipping the answer in pronoun disambiguation .", "we ach...	[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 0 ] }, { "text": "self - supervised method", "nugget_type": "...	[ "we", "propose", "a", "self", "-", "supervised", "method", "to", "solve", "pronoun", "disambiguation", "and", "winograd", "schema", "challenge", "problems", ".", "our", "approach", "exploits", "the", "characteristic", "structure", "of", "training", "corpora", "rel...
ACL	The Dialogue Dodecathlon: Open-Domain Knowledge and Image Grounded Conversational Agents	We introduce dodecaDialogue: a set of 12 tasks that measures if a conversational agent can communicate engagingly with personality and empathy, ask questions, answer questions by utilizing knowledge resources, discuss topics and situations, and perceive and converse about images. By multi-tasking on such a broad large-...	b601d999f30b81efaef652d89d0f28d3	2,020	[ "we introduce dodecadialogue : a set of 12 tasks that measures if a conversational agent can communicate engagingly with personality and empathy , ask questions , answer questions by utilizing knowledge resources , discuss topics and situations , and perceive and converse about images .", "by multi - tasking on s...	[ { "event_type": "CMP", "arguments": [ { "text": "bert pre - trained baseline", "nugget_type": "APP", "argument_type": "Arg2", "tokens": [ "bert", "pre", "-", "trained", "baseline" ], "offsets": [ ...	[ "we", "introduce", "dodecadialogue", ":", "a", "set", "of", "12", "tasks", "that", "measures", "if", "a", "conversational", "agent", "can", "communicate", "engagingly", "with", "personality", "and", "empathy", ",", "ask", "questions", ",", "answer", "questions",...
ACL	Sub-Word Alignment is Still Useful: A Vest-Pocket Method for Enhancing Low-Resource Machine Translation	We leverage embedding duplication between aligned sub-words to extend the Parent-Child transfer learning method, so as to improve low-resource machine translation. We conduct experiments on benchmark datasets of My-En, Id-En and Tr-En translation scenarios. The test results show that our method produces substantial imp...	f40e67b3590f81a7b0f8005f63c0d103	2,022	[ "we leverage embedding duplication between aligned sub - words to extend the parent - child transfer learning method , so as to improve low - resource machine translation .", "we conduct experiments on benchmark datasets of my - en , id - en and tr - en translation scenarios .", "the test results show that our ...	[ { "event_type": "MDS", "arguments": [ { "text": "improve", "nugget_type": "E-PUR", "argument_type": "Target", "tokens": [ "improve" ], "offsets": [ 22 ] }, { "text": "embedding duplication between aligned s...	[ "we", "leverage", "embedding", "duplication", "between", "aligned", "sub", "-", "words", "to", "extend", "the", "parent", "-", "child", "transfer", "learning", "method", ",", "so", "as", "to", "improve", "low", "-", "resource", "machine", "translation", ".", ...
ACL	Parallel Corpus Filtering via Pre-trained Language Models	Web-crawled data provides a good source of parallel corpora for training machine translation models. It is automatically obtained, but extremely noisy, and recent work shows that neural machine translation systems are more sensitive to noise than traditional statistical machine translation methods. In this paper, we pr...	03f63f7d2a1cbc816860924bd533888f	2,020	[ "web - crawled data provides a good source of parallel corpora for training machine translation models .", "it is automatically obtained , but extremely noisy , and recent work shows that neural machine translation systems are more sensitive to noise than traditional statistical machine translation methods .", ...	[ { "event_type": "ITT", "arguments": [ { "text": "web - crawled data", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "web", "-", "crawled", "data" ], "offsets": [ 0, 1, 2...	[ "web", "-", "crawled", "data", "provides", "a", "good", "source", "of", "parallel", "corpora", "for", "training", "machine", "translation", "models", ".", "it", "is", "automatically", "obtained", ",", "but", "extremely", "noisy", ",", "and", "recent", "work", ...
ACL	Synthetic QA Corpora Generation with Roundtrip Consistency	We introduce a novel method of generating synthetic question answering corpora by combining models of question generation and answer extraction, and by filtering the results to ensure roundtrip consistency. By pretraining on the resulting corpora we obtain significant improvements on SQuAD2 and NQ, establishing a new s...	54c8198933fc690451f9eafd8e1d9efe	2,019	[ "we introduce a novel method of generating synthetic question answering corpora by combining models of question generation and answer extraction , and by filtering the results to ensure roundtrip consistency .", "by pretraining on the resulting corpora we obtain significant improvements on squad2 and nq , establi...	[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 0 ] }, { "text": "generating", "nugget_type": "E-PUR", ...	[ "we", "introduce", "a", "novel", "method", "of", "generating", "synthetic", "question", "answering", "corpora", "by", "combining", "models", "of", "question", "generation", "and", "answer", "extraction", ",", "and", "by", "filtering", "the", "results", "to", "en...
ACL	Continual Few-shot Relation Learning via Embedding Space Regularization and Data Augmentation	Existing continual relation learning (CRL) methods rely on plenty of labeled training data for learning a new task, which can be hard to acquire in real scenario as getting large and representative labeled data is often expensive and time-consuming. It is therefore necessary for the model to learn novel relational patt...	06e07ff498e7c6ed929109ab657bbd6a	2,022	[ "existing continual relation learning ( crl ) methods rely on plenty of labeled training data for learning a new task , which can be hard to acquire in real scenario as getting large and representative labeled data is often expensive and time - consuming .", "it is therefore necessary for the model to learn novel...	[ { "event_type": "ITT", "arguments": [ { "text": "continual relation learning methods", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "continual", "relation", "learning", "methods" ], "offsets": [ ...	[ "existing", "continual", "relation", "learning", "(", "crl", ")", "methods", "rely", "on", "plenty", "of", "labeled", "training", "data", "for", "learning", "a", "new", "task", ",", "which", "can", "be", "hard", "to", "acquire", "in", "real", "scenario", "...

ACL

A Lightweight Recurrent Network for Sequence Modeling

Recurrent networks have achieved great success on various sequential tasks with the assistance of complex recurrent units, but suffer from severe computational inefficiency due to weak parallelization. One direction to alleviate this issue is to shift heavy computations outside the recurrence. In this paper, we propose...

e7d6faf5905e216a5d25cc8404b063c9

2,019

[ "recurrent networks have achieved great success on various sequential tasks with the assistance of complex recurrent units , but suffer from severe computational inefficiency due to weak parallelization .", "one direction to alleviate this issue is to shift heavy computations outside the recurrence .", "in this...

[ { "event_type": "ITT", "arguments": [ { "text": "sequential tasks", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "sequential", "tasks" ], "offsets": [ 8, 9 ] } ], "trigger": { ...

[ "recurrent", "networks", "have", "achieved", "great", "success", "on", "various", "sequential", "tasks", "with", "the", "assistance", "of", "complex", "recurrent", "units", ",", "but", "suffer", "from", "severe", "computational", "inefficiency", "due", "to", "weak...

ACL

Contrastive Self-Supervised Learning for Commonsense Reasoning

We propose a self-supervised method to solve Pronoun Disambiguation and Winograd Schema Challenge problems. Our approach exploits the characteristic structure of training corpora related to so-called “trigger” words, which are responsible for flipping the answer in pronoun disambiguation. We achieve such commonsense re...

67aee4c4326695393662a0a4c6051845

2,020

[ "we propose a self - supervised method to solve pronoun disambiguation and winograd schema challenge problems .", "our approach exploits the characteristic structure of training corpora related to so - called “ trigger ” words , which are responsible for flipping the answer in pronoun disambiguation .", "we ach...

[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 0 ] }, { "text": "self - supervised method", "nugget_type": "...

[ "we", "propose", "a", "self", "-", "supervised", "method", "to", "solve", "pronoun", "disambiguation", "and", "winograd", "schema", "challenge", "problems", ".", "our", "approach", "exploits", "the", "characteristic", "structure", "of", "training", "corpora", "rel...

ACL

The Dialogue Dodecathlon: Open-Domain Knowledge and Image Grounded Conversational Agents

We introduce dodecaDialogue: a set of 12 tasks that measures if a conversational agent can communicate engagingly with personality and empathy, ask questions, answer questions by utilizing knowledge resources, discuss topics and situations, and perceive and converse about images. By multi-tasking on such a broad large-...

b601d999f30b81efaef652d89d0f28d3

2,020

[ "we introduce dodecadialogue : a set of 12 tasks that measures if a conversational agent can communicate engagingly with personality and empathy , ask questions , answer questions by utilizing knowledge resources , discuss topics and situations , and perceive and converse about images .", "by multi - tasking on s...

[ { "event_type": "CMP", "arguments": [ { "text": "bert pre - trained baseline", "nugget_type": "APP", "argument_type": "Arg2", "tokens": [ "bert", "pre", "-", "trained", "baseline" ], "offsets": [ ...

[ "we", "introduce", "dodecadialogue", ":", "a", "set", "of", "12", "tasks", "that", "measures", "if", "a", "conversational", "agent", "can", "communicate", "engagingly", "with", "personality", "and", "empathy", ",", "ask", "questions", ",", "answer", "questions",...

ACL

Sub-Word Alignment is Still Useful: A Vest-Pocket Method for Enhancing Low-Resource Machine Translation

We leverage embedding duplication between aligned sub-words to extend the Parent-Child transfer learning method, so as to improve low-resource machine translation. We conduct experiments on benchmark datasets of My-En, Id-En and Tr-En translation scenarios. The test results show that our method produces substantial imp...

f40e67b3590f81a7b0f8005f63c0d103

2,022

[ "we leverage embedding duplication between aligned sub - words to extend the parent - child transfer learning method , so as to improve low - resource machine translation .", "we conduct experiments on benchmark datasets of my - en , id - en and tr - en translation scenarios .", "the test results show that our ...

[ { "event_type": "MDS", "arguments": [ { "text": "improve", "nugget_type": "E-PUR", "argument_type": "Target", "tokens": [ "improve" ], "offsets": [ 22 ] }, { "text": "embedding duplication between aligned s...

[ "we", "leverage", "embedding", "duplication", "between", "aligned", "sub", "-", "words", "to", "extend", "the", "parent", "-", "child", "transfer", "learning", "method", ",", "so", "as", "to", "improve", "low", "-", "resource", "machine", "translation", ".", ...

ACL

Parallel Corpus Filtering via Pre-trained Language Models

Web-crawled data provides a good source of parallel corpora for training machine translation models. It is automatically obtained, but extremely noisy, and recent work shows that neural machine translation systems are more sensitive to noise than traditional statistical machine translation methods. In this paper, we pr...

03f63f7d2a1cbc816860924bd533888f

2,020

[ "web - crawled data provides a good source of parallel corpora for training machine translation models .", "it is automatically obtained , but extremely noisy , and recent work shows that neural machine translation systems are more sensitive to noise than traditional statistical machine translation methods .", ...

[ { "event_type": "ITT", "arguments": [ { "text": "web - crawled data", "nugget_type": "TAK", "argument_type": "Target", "tokens": [ "web", "-", "crawled", "data" ], "offsets": [ 0, 1, 2...

[ "web", "-", "crawled", "data", "provides", "a", "good", "source", "of", "parallel", "corpora", "for", "training", "machine", "translation", "models", ".", "it", "is", "automatically", "obtained", ",", "but", "extremely", "noisy", ",", "and", "recent", "work", ...

ACL

Synthetic QA Corpora Generation with Roundtrip Consistency

We introduce a novel method of generating synthetic question answering corpora by combining models of question generation and answer extraction, and by filtering the results to ensure roundtrip consistency. By pretraining on the resulting corpora we obtain significant improvements on SQuAD2 and NQ, establishing a new s...

54c8198933fc690451f9eafd8e1d9efe

2,019

[ "we introduce a novel method of generating synthetic question answering corpora by combining models of question generation and answer extraction , and by filtering the results to ensure roundtrip consistency .", "by pretraining on the resulting corpora we obtain significant improvements on squad2 and nq , establi...

[ { "event_type": "PRP", "arguments": [ { "text": "we", "nugget_type": "OG", "argument_type": "Proposer", "tokens": [ "we" ], "offsets": [ 0 ] }, { "text": "generating", "nugget_type": "E-PUR", ...

[ "we", "introduce", "a", "novel", "method", "of", "generating", "synthetic", "question", "answering", "corpora", "by", "combining", "models", "of", "question", "generation", "and", "answer", "extraction", ",", "and", "by", "filtering", "the", "results", "to", "en...

ACL

Continual Few-shot Relation Learning via Embedding Space Regularization and Data Augmentation

Existing continual relation learning (CRL) methods rely on plenty of labeled training data for learning a new task, which can be hard to acquire in real scenario as getting large and representative labeled data is often expensive and time-consuming. It is therefore necessary for the model to learn novel relational patt...

06e07ff498e7c6ed929109ab657bbd6a

2,022

[ "existing continual relation learning ( crl ) methods rely on plenty of labeled training data for learning a new task , which can be hard to acquire in real scenario as getting large and representative labeled data is often expensive and time - consuming .", "it is therefore necessary for the model to learn novel...

[ { "event_type": "ITT", "arguments": [ { "text": "continual relation learning methods", "nugget_type": "APP", "argument_type": "Target", "tokens": [ "continual", "relation", "learning", "methods" ], "offsets": [ ...

[ "existing", "continual", "relation", "learning", "(", "crl", ")", "methods", "rely", "on", "plenty", "of", "labeled", "training", "data", "for", "learning", "a", "new", "task", ",", "which", "can", "be", "hard", "to", "acquire", "in", "real", "scenario", "...