paper_id
stringlengths
17
19
title
stringlengths
27
144
paper_url
stringlengths
43
45
authors
listlengths
1
37
abstract
large_stringlengths
396
1.69k
anthology_id
stringlengths
17
19
doi
stringlengths
29
31
award
stringclasses
0 values
arxiv_id
stringlengths
10
10
arxiv_id_source
stringclasses
2 values
2021.emnlp-main.501
Data Collection vs. Knowledge Graph Completion: What is Needed to Improve Coverage?
https://aclanthology.org/2021.emnlp-main.501/
[ "Kenneth Church", "Yuchen Bian" ]
This survey/position paper discusses ways to improve coverage of resources such as WordNet. Rapp estimated correlations, rho, between corpus statistics and pyscholinguistic norms. rho improves with quantity (corpus size) and quality (balance). 1M words is enough for simple estimates (unigram frequencies), but at least ...
2021.emnlp-main.501
10.18653/v1/2021.emnlp-main.501
null
null
null
2021.emnlp-main.502
Universal Sentence Representation Learning with Conditional Masked Language Model
https://aclanthology.org/2021.emnlp-main.502/
[ "Ziyi Yang", "Yinfei Yang", "Daniel Cer", "Jax Law", "Eric Darve" ]
This paper presents a novel training method, Conditional Masked Language Modeling (CMLM), to effectively learn sentence representations on large scale unlabeled corpora. CMLM integrates sentence representation learning into MLM training by conditioning on the encoded vectors of adjacent sentences. Our English CMLM mode...
2021.emnlp-main.502
10.18653/v1/2021.emnlp-main.502
null
2012.14388
title_snapshot
2021.emnlp-main.503
On the Benefit of Syntactic Supervision for Cross-lingual Transfer in Semantic Role Labeling
https://aclanthology.org/2021.emnlp-main.503/
[ "Zhisong Zhang", "Emma Strubell", "Eduard Hovy" ]
Although recent developments in neural architectures and pre-trained representations have greatly increased state-of-the-art model performance on fully-supervised semantic role labeling (SRL), the task remains challenging for languages where supervised SRL training data are not abundant. Cross-lingual learning can impr...
2021.emnlp-main.503
10.18653/v1/2021.emnlp-main.503
null
null
null
2021.emnlp-main.504
Implicit Premise Generation with Discourse-aware Commonsense Knowledge Models
https://aclanthology.org/2021.emnlp-main.504/
[ "Tuhin Chakrabarty", "Aadit Trivedi", "Smaranda Muresan" ]
Enthymemes are defined as arguments where a premise or conclusion is left implicit. We tackle the task of generating the implicit premise in an enthymeme, which requires not only an understanding of the stated conclusion and premise but also additional inferences that could depend on commonsense knowledge. The largest ...
2021.emnlp-main.504
10.18653/v1/2021.emnlp-main.504
null
2109.05358
title_snapshot
2021.emnlp-main.505
Inducing Transformer’s Compositional Generalization Ability via Auxiliary Sequence Prediction Tasks
https://aclanthology.org/2021.emnlp-main.505/
[ "Yichen Jiang", "Mohit Bansal" ]
Systematic compositionality is an essential mechanism in human language, allowing the recombination of known parts to create novel expressions. However, existing neural models have been shown to lack this basic ability in learning symbolic structures. Motivated by the failure of a Transformer model on the SCAN composit...
2021.emnlp-main.505
10.18653/v1/2021.emnlp-main.505
null
2109.15256
title_snapshot
2021.emnlp-main.506
Flexible Generation of Natural Language Deductions
https://aclanthology.org/2021.emnlp-main.506/
[ "Kaj Bostrom", "Xinyu Zhao", "Swarat Chaudhuri", "Greg Durrett" ]
An interpretable system for open-domain reasoning needs to express its reasoning process in a transparent form. Natural language is an attractive representation for this purpose — it is both highly expressive and easy for humans to understand. However, manipulating natural language statements in logically consistent wa...
2021.emnlp-main.506
10.18653/v1/2021.emnlp-main.506
null
2104.08825
title_snapshot
2021.emnlp-main.507
Structure-aware Fine-tuning of Sequence-to-sequence Transformers for Transition-based AMR Parsing
https://aclanthology.org/2021.emnlp-main.507/
[ "Jiawei Zhou", "Tahira Naseem", "Ramón Fernandez Astudillo", "Young-Suk Lee", "Radu Florian", "Salim Roukos" ]
Predicting linearized Abstract Meaning Representation (AMR) graphs using pre-trained sequence-to-sequence Transformer models has recently led to large improvements on AMR parsing benchmarks. These parsers are simple and avoid explicit modeling of structure but lack desirable properties such as graph well-formedness gua...
2021.emnlp-main.507
10.18653/v1/2021.emnlp-main.507
null
2110.15534
title_snapshot
2021.emnlp-main.508
Think about it! Improving defeasible reasoning by first modeling the question scenario.
https://aclanthology.org/2021.emnlp-main.508/
[ "Aman Madaan", "Niket Tandon", "Dheeraj Rajagopal", "Peter Clark", "Yiming Yang", "Eduard Hovy" ]
Defeasible reasoning is the mode of reasoning where conclusions can be overturned by taking into account new evidence. Existing cognitive science literature on defeasible reasoning suggests that a person forms a “mental model” of the problem scenario before answering questions. Our research goal asks whether neural mod...
2021.emnlp-main.508
10.18653/v1/2021.emnlp-main.508
null
2110.12349
title_snapshot
2021.emnlp-main.509
Open Aspect Target Sentiment Classification with Natural Language Prompts
https://aclanthology.org/2021.emnlp-main.509/
[ "Ronald Seoh", "Ian Birle", "Mrinal Tak", "Haw-Shiuan Chang", "Brian Pinette", "Alfred Hough" ]
For many business applications, we often seek to analyze sentiments associated with any arbitrary aspects of commercial products, despite having a very limited amount of labels or even without any labels at all. However, existing aspect target sentiment classification (ATSC) models are not trainable if annotated datase...
2021.emnlp-main.509
10.18653/v1/2021.emnlp-main.509
null
2109.03685
title_snapshot
2021.emnlp-main.510
Does BERT Learn as Humans Perceive? Understanding Linguistic Styles through Lexica
https://aclanthology.org/2021.emnlp-main.510/
[ "Shirley Anugrah Hayati", "Dongyeop Kang", "Lyle Ungar" ]
People convey their intention and attitude through linguistic styles of the text that they write. In this study, we investigate lexicon usages across styles throughout two lenses: human perception and machine word importance, since words differ in the strength of the stylistic cues that they provide. To collect labels ...
2021.emnlp-main.510
10.18653/v1/2021.emnlp-main.510
null
2109.02738
title_snapshot
2021.emnlp-main.511
Improving Stance Detection with Multi-Dataset Learning and Knowledge Distillation
https://aclanthology.org/2021.emnlp-main.511/
[ "Yingjie Li", "Chenye Zhao", "Cornelia Caragea" ]
Stance detection determines whether the author of a text is in favor of, against or neutral to a specific target and provides valuable insights into important events such as legalization of abortion. Despite significant progress on this task, one of the remaining challenges is the scarcity of annotations. Besides, most...
2021.emnlp-main.511
10.18653/v1/2021.emnlp-main.511
null
null
null
2021.emnlp-main.512
Discovering the Unknown Knowns: Turning Implicit Knowledge in the Dataset into Explicit Training Examples for Visual Question Answering
https://aclanthology.org/2021.emnlp-main.512/
[ "Jihyung Kil", "Cheng Zhang", "Dong Xuan", "Wei-Lun Chao" ]
Visual question answering (VQA) is challenging not only because the model has to handle multi-modal information, but also because it is just so hard to collect sufficient training examples — there are too many questions one can ask about an image. As a result, a VQA model trained solely on human-annotated examples coul...
2021.emnlp-main.512
10.18653/v1/2021.emnlp-main.512
null
2109.06122
title_snapshot
2021.emnlp-main.513
Improving Pre-trained Vision-and-Language Embeddings for Phrase Grounding
https://aclanthology.org/2021.emnlp-main.513/
[ "Zi-Yi Dou", "Nanyun Peng" ]
Phrase grounding aims to map textual phrases to their associated image regions, which can be a prerequisite for multimodal reasoning and can benefit tasks requiring identifying objects based on language. With pre-trained vision-and-language models achieving impressive performance across tasks, it remains unclear if we ...
2021.emnlp-main.513
10.18653/v1/2021.emnlp-main.513
null
null
null
2021.emnlp-main.514
Sequential Randomized Smoothing for Adversarially Robust Speech Recognition
https://aclanthology.org/2021.emnlp-main.514/
[ "Raphael Olivier", "Bhiksha Raj" ]
While Automatic Speech Recognition has been shown to be vulnerable to adversarial attacks, defenses against these attacks are still lagging. Existing, naive defenses can be partially broken with an adaptive attack. In classification tasks, the Randomized Smoothing paradigm has been shown to be effective at defending mo...
2021.emnlp-main.514
10.18653/v1/2021.emnlp-main.514
null
2112.03000
title_snapshot
2021.emnlp-main.515
Hitting your MARQ: Multimodal ARgument Quality Assessment in Long Debate Video
https://aclanthology.org/2021.emnlp-main.515/
[ "Md Kamrul Hasan", "James Spann", "Masum Hasan", "Md Saiful Islam", "Kurtis Haut", "Rada Mihalcea", "Ehsan Hoque" ]
The combination of gestures, intonations, and textual content plays a key role in argument delivery. However, the current literature mostly considers textual content while assessing the quality of an argument, and it is limited to datasets containing short sequences (18-48 words). In this paper, we study argument quali...
2021.emnlp-main.515
10.18653/v1/2021.emnlp-main.515
null
null
null
2021.emnlp-main.516
Mind the Context: The Impact of Contextualization in Neural Module Networks for Grounding Visual Referring Expressions
https://aclanthology.org/2021.emnlp-main.516/
[ "Arjun Akula", "Spandana Gella", "Keze Wang", "Song-Chun Zhu", "Siva Reddy" ]
Neural module networks (NMN) are a popular approach for grounding visual referring expressions. Prior implementations of NMN use pre-defined and fixed textual inputs in their module instantiation. This necessitates a large number of modules as they lack the ability to share weights and exploit associations between simi...
2021.emnlp-main.516
10.18653/v1/2021.emnlp-main.516
null
null
null
2021.emnlp-main.517
Weakly-Supervised Visual-Retriever-Reader for Knowledge-based Question Answering
https://aclanthology.org/2021.emnlp-main.517/
[ "Man Luo", "Yankai Zeng", "Pratyay Banerjee", "Chitta Baral" ]
Knowledge-based visual question answering (VQA) requires answering questions with external knowledge in addition to the content of images. One dataset that is mostly used in evaluating knowledge-based VQA is OK-VQA, but it lacks a gold standard knowledge corpus for retrieval. Existing work leverage different knowledge ...
2021.emnlp-main.517
10.18653/v1/2021.emnlp-main.517
null
2109.04014
title_snapshot
2021.emnlp-main.518
NDH-Full: Learning and Evaluating Navigational Agents on Full-Length Dialogue
https://aclanthology.org/2021.emnlp-main.518/
[ "Hyounghun Kim", "Jialu Li", "Mohit Bansal" ]
Communication between human and mobile agents is getting increasingly important as such agents are widely deployed in our daily lives. Vision-and-Dialogue Navigation is one of the tasks that evaluate the agent’s ability to interact with humans for assistance and navigate based on natural language responses. In this pap...
2021.emnlp-main.518
10.18653/v1/2021.emnlp-main.518
null
null
null
2021.emnlp-main.519
Timeline Summarization based on Event Graph Compression via Time-Aware Optimal Transport
https://aclanthology.org/2021.emnlp-main.519/
[ "Manling Li", "Tengfei Ma", "Mo Yu", "Lingfei Wu", "Tian Gao", "Heng Ji", "Kathleen McKeown" ]
Timeline Summarization identifies major events from a news collection and describes them following temporal order, with key dates tagged. Previous methods generally generate summaries separately for each date after they determine the key dates of events. These methods overlook the events’ intra-structures (arguments) a...
2021.emnlp-main.519
10.18653/v1/2021.emnlp-main.519
null
null
null
2021.emnlp-main.520
StreamHover: Livestream Transcript Summarization and Annotation
https://aclanthology.org/2021.emnlp-main.520/
[ "Sangwoo Cho", "Franck Dernoncourt", "Tim Ganter", "Trung Bui", "Nedim Lipka", "Walter Chang", "Hailin Jin", "Jonathan Brandt", "Hassan Foroosh", "Fei Liu" ]
With the explosive growth of livestream broadcasting, there is an urgent need for new summarization technology that enables us to create a preview of streamed content and tap into this wealth of knowledge. However, the problem is nontrivial due to the informal nature of spoken language. Further, there has been a shorta...
2021.emnlp-main.520
10.18653/v1/2021.emnlp-main.520
null
2109.05160
title_snapshot
2021.emnlp-main.521
Cross-Register Projection for Headline Part of Speech Tagging
https://aclanthology.org/2021.emnlp-main.521/
[ "Adrian Benton", "Hanyang Li", "Igor Malioutov" ]
Part of speech (POS) tagging is a familiar NLP task. State of the art taggers routinely achieve token-level accuracies of over 97% on news body text, evidence that the problem is well understood. However, the register of English news headlines, “headlinese”, is very different from the register of long-form text, causin...
2021.emnlp-main.521
10.18653/v1/2021.emnlp-main.521
null
2109.07483
title_snapshot
2021.emnlp-main.522
Editing Factual Knowledge in Language Models
https://aclanthology.org/2021.emnlp-main.522/
[ "Nicola De Cao", "Wilker Aziz", "Ivan Titov" ]
The factual knowledge acquired during pre-training and stored in the parameters of Language Models (LMs) can be useful in downstream tasks (e.g., question answering or textual inference). However, some facts can be incorrectly induced or become obsolete over time. We present KnowledgeEditor, a method which can be used ...
2021.emnlp-main.522
10.18653/v1/2021.emnlp-main.522
null
2104.08164
title_snapshot
2021.emnlp-main.523
Sparse Attention with Linear Units
https://aclanthology.org/2021.emnlp-main.523/
[ "Biao Zhang", "Ivan Titov", "Rico Sennrich" ]
Recently, it has been argued that encoder-decoder models can be made more interpretable by replacing the softmax function in the attention with its sparse variants. In this work, we introduce a novel, simple method for achieving sparsity in attention: we replace the softmax activation with a ReLU, and show that sparsit...
2021.emnlp-main.523
10.18653/v1/2021.emnlp-main.523
null
2104.07012
title_snapshot
2021.emnlp-main.524
Knowledge Base Completion Meets Transfer Learning
https://aclanthology.org/2021.emnlp-main.524/
[ "Vid Kocijan", "Thomas Lukasiewicz" ]
The aim of knowledge base completion is to predict unseen facts from existing facts in knowledge bases. In this work, we introduce the first approach for transfer of knowledge from one collection of facts to another without the need for entity or relation matching. The method works for both canonicalized knowledge base...
2021.emnlp-main.524
10.18653/v1/2021.emnlp-main.524
null
2108.13073
title_snapshot
2021.emnlp-main.525
SPECTRA: Sparse Structured Text Rationalization
https://aclanthology.org/2021.emnlp-main.525/
[ "Nuno M. Guerreiro", "André F. T. Martins" ]
Selective rationalization aims to produce decisions along with rationales (e.g., text highlights or word alignments between two sentences). Commonly, rationales are modeled as stochastic binary masks, requiring sampling-based gradient estimators, which complicates training and requires careful hyperparameter tuning. Sp...
2021.emnlp-main.525
10.18653/v1/2021.emnlp-main.525
null
2109.04552
title_snapshot
2021.emnlp-main.526
Towards Zero-Shot Knowledge Distillation for Natural Language Processing
https://aclanthology.org/2021.emnlp-main.526/
[ "Ahmad Rashid", "Vasileios Lioutas", "Abbas Ghaddar", "Mehdi Rezagholizadeh" ]
Knowledge distillation (KD) is a common knowledge transfer algorithm used for model compression across a variety of deep learning based natural language processing (NLP) solutions. In its regular manifestations, KD requires access to the teacher’s training data for knowledge transfer to the student network. However, pr...
2021.emnlp-main.526
10.18653/v1/2021.emnlp-main.526
null
2012.15495
title_snapshot
2021.emnlp-main.527
Adversarial Regularization as Stackelberg Game: An Unrolled Optimization Approach
https://aclanthology.org/2021.emnlp-main.527/
[ "Simiao Zuo", "Chen Liang", "Haoming Jiang", "Xiaodong Liu", "Pengcheng He", "Jianfeng Gao", "Weizhu Chen", "Tuo Zhao" ]
Adversarial regularization has been shown to improve the generalization performance of deep learning models in various natural language processing tasks. Existing works usually formulate the method as a zero-sum game, which is solved by alternating gradient descent/ascent algorithms. Such a formulation treats the adver...
2021.emnlp-main.527
10.18653/v1/2021.emnlp-main.527
null
2104.04886
title_snapshot
2021.emnlp-main.528
Aspect-Controllable Opinion Summarization
https://aclanthology.org/2021.emnlp-main.528/
[ "Reinald Kim Amplayo", "Stefanos Angelidis", "Mirella Lapata" ]
Recent work on opinion summarization produces general summaries based on a set of input reviews and the popularity of opinions expressed in them. In this paper, we propose an approach that allows the generation of customized summaries based on aspect queries (e.g., describing the location and room of a hotel). Using a ...
2021.emnlp-main.528
10.18653/v1/2021.emnlp-main.528
null
2109.03171
title_snapshot
2021.emnlp-main.529
QuestEval: Summarization Asks for Fact-based Evaluation
https://aclanthology.org/2021.emnlp-main.529/
[ "Thomas Scialom", "Paul-Alexis Dray", "Sylvain Lamprier", "Benjamin Piwowarski", "Jacopo Staiano", "Alex Wang", "Patrick Gallinari" ]
Summarization evaluation remains an open research problem: current metrics such as ROUGE are known to be limited and to correlate poorly with human judgments. To alleviate this issue, recent work has proposed evaluation metrics which rely on question answering models to assess whether a summary contains all the relevan...
2021.emnlp-main.529
10.18653/v1/2021.emnlp-main.529
null
2103.12693
title_snapshot
2021.emnlp-main.530
Simple Conversational Data Augmentation for Semi-supervised Abstractive Dialogue Summarization
https://aclanthology.org/2021.emnlp-main.530/
[ "Jiaao Chen", "Diyi Yang" ]
Abstractive conversation summarization has received growing attention while most current state-of-the-art summarization models heavily rely on human-annotated summaries. To reduce the dependence on labeled summaries, in this work, we present a simple yet effective set of Conversational Data Augmentation (CODA) methods ...
2021.emnlp-main.530
10.18653/v1/2021.emnlp-main.530
null
null
null
2021.emnlp-main.531
Finding a Balanced Degree of Automation for Summary Evaluation
https://aclanthology.org/2021.emnlp-main.531/
[ "Shiyue Zhang", "Mohit Bansal" ]
Human evaluation for summarization tasks is reliable but brings in issues of reproducibility and high costs. Automatic metrics are cheap and reproducible but sometimes poorly correlated with human judgment. In this work, we propose flexible semiautomatic to automatic summary evaluation metrics, following the Pyramid hu...
2021.emnlp-main.531
10.18653/v1/2021.emnlp-main.531
null
2109.11503
title_snapshot
2021.emnlp-main.532
CLIFF: Contrastive Learning for Improving Faithfulness and Factuality in Abstractive Summarization
https://aclanthology.org/2021.emnlp-main.532/
[ "Shuyang Cao", "Lu Wang" ]
We study generating abstractive summaries that are faithful and factually consistent with the given articles. A novel contrastive learning formulation is presented, which leverages both reference summaries, as positive training data, and automatically generated erroneous summaries, as negative training data, to train s...
2021.emnlp-main.532
10.18653/v1/2021.emnlp-main.532
null
2109.09209
title_snapshot
2021.emnlp-main.533
Multilingual Unsupervised Neural Machine Translation with Denoising Adapters
https://aclanthology.org/2021.emnlp-main.533/
[ "Ahmet Üstün", "Alexandre Berard", "Laurent Besacier", "Matthias Gallé" ]
We consider the problem of multilingual unsupervised machine translation, translating to and from languages that only have monolingual data by using auxiliary parallel language pairs. For this problem the standard procedure so far to leverage the monolingual data is _back-translation_, which is computationally costly a...
2021.emnlp-main.533
10.18653/v1/2021.emnlp-main.533
null
2110.10472
title_snapshot
2021.emnlp-main.534
BERT, mBERT, or BiBERT? A Study on Contextualized Embeddings for Neural Machine Translation
https://aclanthology.org/2021.emnlp-main.534/
[ "Haoran Xu", "Benjamin Van Durme", "Kenton Murray" ]
The success of bidirectional encoders using masked language models, such as BERT, on numerous natural language processing tasks has prompted researchers to attempt to incorporate these pre-trained models into neural machine translation (NMT) systems. However, proposed methods for incorporating pre-trained models are no...
2021.emnlp-main.534
10.18653/v1/2021.emnlp-main.534
null
2109.04588
title_snapshot
2021.emnlp-main.535
Controlling Machine Translation for Multiple Attributes with Additive Interventions
https://aclanthology.org/2021.emnlp-main.535/
[ "Andrea Schioppa", "David Vilar", "Artem Sokolov", "Katja Filippova" ]
Fine-grained control of machine translation (MT) outputs along multiple attributes is critical for many modern MT applications and is a requirement for gaining users’ trust. A standard approach for exerting control in MT is to prepend the input with a special tag to signal the desired output attribute. Despite its simp...
2021.emnlp-main.535
10.18653/v1/2021.emnlp-main.535
null
null
null
2021.emnlp-main.536
A Generative Framework for Simultaneous Machine Translation
https://aclanthology.org/2021.emnlp-main.536/
[ "Yishu Miao", "Phil Blunsom", "Lucia Specia" ]
We propose a generative framework for simultaneous machine translation. Conventional approaches use a fixed number of source words to translate or learn dynamic policies for the number of source words by reinforcement learning. Here we formulate simultaneous translation as a structural sequence-to-sequence learning pro...
2021.emnlp-main.536
10.18653/v1/2021.emnlp-main.536
null
null
null
2021.emnlp-main.537
It Is Not As Good As You Think! Evaluating Simultaneous Machine Translation on Interpretation Data
https://aclanthology.org/2021.emnlp-main.537/
[ "Jinming Zhao", "Philip Arthur", "Gholamreza Haffari", "Trevor Cohn", "Ehsan Shareghi" ]
Most existing simultaneous machine translation (SiMT) systems are trained and evaluated on offline translation corpora. We argue that SiMT systems should be trained and tested on real interpretation data. To illustrate this argument, we propose an interpretation test set and conduct a realistic evaluation of SiMT train...
2021.emnlp-main.537
10.18653/v1/2021.emnlp-main.537
null
2110.05213
title_snapshot
2021.emnlp-main.538
Boosting Cross-Lingual Transfer via Self-Learning with Uncertainty Estimation
https://aclanthology.org/2021.emnlp-main.538/
[ "Liyan Xu", "Xuchao Zhang", "Xujiang Zhao", "Haifeng Chen", "Feng Chen", "Jinho D. Choi" ]
Recent multilingual pre-trained language models have achieved remarkable zero-shot performance, where the model is only finetuned on one source language and directly evaluated on target languages. In this work, we propose a self-learning framework that further utilizes unlabeled data of target languages, combined with ...
2021.emnlp-main.538
10.18653/v1/2021.emnlp-main.538
null
2109.00194
title_snapshot
2021.emnlp-main.539
Levenshtein Training for Word-level Quality Estimation
https://aclanthology.org/2021.emnlp-main.539/
[ "Shuoyang Ding", "Marcin Junczys-Dowmunt", "Matt Post", "Philipp Koehn" ]
We propose a novel scheme to use the Levenshtein Transformer to perform the task of word-level quality estimation. A Levenshtein Transformer is a natural fit for this task: trained to perform decoding in an iterative manner, a Levenshtein Transformer can learn to post-edit without explicit supervision. To further minim...
2021.emnlp-main.539
10.18653/v1/2021.emnlp-main.539
null
2109.05611
title_snapshot
2021.emnlp-main.540
Interactive Machine Comprehension with Dynamic Knowledge Graphs
https://aclanthology.org/2021.emnlp-main.540/
[ "Xingdi Yuan" ]
Interactive machine reading comprehension (iMRC) is machine comprehension tasks where knowledge sources are partially observable. An agent must interact with an environment sequentially to gather necessary knowledge in order to answer a question. We hypothesize that graph representations are good inductive biases, whic...
2021.emnlp-main.540
10.18653/v1/2021.emnlp-main.540
null
2109.00077
title_snapshot
2021.emnlp-main.541
Residual Adapters for Parameter-Efficient ASR Adaptation to Atypical and Accented Speech
https://aclanthology.org/2021.emnlp-main.541/
[ "Katrin Tomanek", "Vicky Zayats", "Dirk Padfield", "Kara Vaillancourt", "Fadi Biadsy" ]
Automatic Speech Recognition (ASR) systems are often optimized to work best for speakers with canonical speech patterns. Unfortunately, these systems perform poorly when tested on atypical speech and heavily accented speech. It has previously been shown that personalization through model fine-tuning substantially impro...
2021.emnlp-main.541
10.18653/v1/2021.emnlp-main.541
null
2109.06952
title_snapshot
2021.emnlp-main.542
Visual News: Benchmark and Challenges in News Image Captioning
https://aclanthology.org/2021.emnlp-main.542/
[ "Fuxiao Liu", "Yinghan Wang", "Tianlu Wang", "Vicente Ordonez" ]
We propose Visual News Captioner, an entity-aware model for the task of news image captioning. We also introduce Visual News, a large-scale benchmark consisting of more than one million news images along with associated news articles, image captions, author information, and other metadata. Unlike the standard image cap...
2021.emnlp-main.542
10.18653/v1/2021.emnlp-main.542
null
2010.03743
title_snapshot
2021.emnlp-main.543
Integrating Visuospatial, Linguistic, and Commonsense Structure into Story Visualization
https://aclanthology.org/2021.emnlp-main.543/
[ "Adyasha Maharana", "Mohit Bansal" ]
While much research has been done in text-to-image synthesis, little work has been done to explore the usage of linguistic structure of the input text. Such information is even more important for story visualization since its inputs have an explicit narrative structure that needs to be translated into an image sequence...
2021.emnlp-main.543
10.18653/v1/2021.emnlp-main.543
null
2110.10834
title_snapshot
2021.emnlp-main.544
VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding
https://aclanthology.org/2021.emnlp-main.544/
[ "Hu Xu", "Gargi Ghosh", "Po-Yao Huang", "Dmytro Okhonko", "Armen Aghajanyan", "Florian Metze", "Luke Zettlemoyer", "Christoph Feichtenhofer" ]
We present VideoCLIP, a contrastive approach to pre-train a unified model for zero-shot video and text understanding, without using any labels on downstream tasks. VideoCLIP trains a transformer for video and text by contrasting temporally overlapping positive video-text pairs with hard negatives from nearest neighbor ...
2021.emnlp-main.544
10.18653/v1/2021.emnlp-main.544
null
2109.14084
title_snapshot
2021.emnlp-main.545
NewsCLIPpings: Automatic Generation of Out-of-Context Multimodal Media
https://aclanthology.org/2021.emnlp-main.545/
[ "Grace Luo", "Trevor Darrell", "Anna Rohrbach" ]
Online misinformation is a prevalent societal issue, with adversaries relying on tools ranging from cheap fakes to sophisticated deep fakes. We are motivated by the threat scenario where an image is used out of context to support a certain narrative. While some prior datasets for detecting image-text inconsistency gene...
2021.emnlp-main.545
10.18653/v1/2021.emnlp-main.545
null
2104.05893
title_snapshot
2021.emnlp-main.546
Powering Comparative Classification with Sentiment Analysis via Domain Adaptive Knowledge Transfer
https://aclanthology.org/2021.emnlp-main.546/
[ "Zeyu Li", "Yilong Qin", "Zihan Liu", "Wei Wang" ]
We study Comparative Preference Classification (CPC) which aims at predicting whether a preference comparison exists between two entities in a given sentence and, if so, which entity is preferred over the other. High-quality CPC models can significantly benefit applications such as comparative question answering and re...
2021.emnlp-main.546
10.18653/v1/2021.emnlp-main.546
null
2109.03819
title_snapshot
2021.emnlp-main.547
Tribrid: Stance Classification with Neural Inconsistency Detection
https://aclanthology.org/2021.emnlp-main.547/
[ "Song Yang", "Jacopo Urbani" ]
We study the problem of performing automatic stance classification on social media with neural architectures such as BERT. Although these architectures deliver impressive results, their level is not yet comparable to the one of humans and they might produce errors that have a significant impact on the downstream task (...
2021.emnlp-main.547
10.18653/v1/2021.emnlp-main.547
null
2109.06508
title_snapshot
2021.emnlp-main.548
SYSML: StYlometry with Structure and Multitask Learning: Implications for Darknet Forum Migrant Analysis
https://aclanthology.org/2021.emnlp-main.548/
[ "Pranav Maneriker", "Yuntian He", "Srinivasan Parthasarathy" ]
Darknet market forums are frequently used to exchange illegal goods and services between parties who use encryption to conceal their identities. The Tor network is used to host these markets, which guarantees additional anonymization from IP and location tracking, making it challenging to link across malicious users us...
2021.emnlp-main.548
10.18653/v1/2021.emnlp-main.548
null
2104.00764
title_snapshot
2021.emnlp-main.549
Few-Shot Emotion Recognition in Conversation with Sequential Prototypical Networks
https://aclanthology.org/2021.emnlp-main.549/
[ "Gaël Guibon", "Matthieu Labeau", "Hélène Flamein", "Luce Lefeuvre", "Chloé Clavel" ]
Several recent studies on dyadic human-human interactions have been done on conversations without specific business objectives. However, many companies might benefit from studies dedicated to more precise environments such as after sales services or customer satisfaction surveys. In this work, we place ourselves in the...
2021.emnlp-main.549
10.18653/v1/2021.emnlp-main.549
null
2109.09366
title_snapshot
2021.emnlp-main.550
CLASSIC: Continual and Contrastive Learning of Aspect Sentiment Classification Tasks
https://aclanthology.org/2021.emnlp-main.550/
[ "Zixuan Ke", "Bing Liu", "Hu Xu", "Lei Shu" ]
This paper studies continual learning (CL) of a sequence of aspect sentiment classification (ASC) tasks in a particular CL setting called domain incremental learning (DIL). Each task is from a different domain or product. The DIL setting is particularly suited to ASC because in testing the system needs not know the tas...
2021.emnlp-main.550
10.18653/v1/2021.emnlp-main.550
null
2112.02714
title_snapshot
2021.emnlp-main.551
Implicit Sentiment Analysis with Event-centered Text Representation
https://aclanthology.org/2021.emnlp-main.551/
[ "Deyu Zhou", "Jianan Wang", "Linhai Zhang", "Yulan He" ]
Implicit sentiment analysis, aiming at detecting the sentiment of a sentence without sentiment words, has become an attractive research topic in recent years. In this paper, we focus on event-centric implicit sentiment analysis that utilizes the sentiment-aware event contained in a sentence to infer its sentiment polar...
2021.emnlp-main.551
10.18653/v1/2021.emnlp-main.551
null
null
null
2021.emnlp-main.552
SimCSE: Simple Contrastive Learning of Sentence Embeddings
https://aclanthology.org/2021.emnlp-main.552/
[ "Tianyu Gao", "Xingcheng Yao", "Danqi Chen" ]
This paper presents SimCSE, a simple contrastive learning framework that greatly advances the state-of-the-art sentence embeddings. We first describe an unsupervised approach, which takes an input sentence and predicts itself in a contrastive objective, with only standard dropout used as noise. This simple method works...
2021.emnlp-main.552
10.18653/v1/2021.emnlp-main.552
null
2104.08821
title_snapshot
2021.emnlp-main.553
When is Wall a Pared and when a Muro?: Extracting Rules Governing Lexical Selection
https://aclanthology.org/2021.emnlp-main.553/
[ "Aditi Chaudhary", "Kayo Yin", "Antonios Anastasopoulos", "Graham Neubig" ]
Learning fine-grained distinctions between vocabulary items is a key challenge in learning a new language. For example, the noun “wall” has different lexical manifestations in Spanish – “pared” refers to an indoor wall while “muro” refers to an outside wall. However, this variety of lexical distinction may not be obvio...
2021.emnlp-main.553
10.18653/v1/2021.emnlp-main.553
null
2109.06014
title_snapshot
2021.emnlp-main.554
Aligning Actions Across Recipe Graphs
https://aclanthology.org/2021.emnlp-main.554/
[ "Lucia Donatelli", "Theresa Schmidt", "Debanjali Biswas", "Arne Köhn", "Fangzhou Zhai", "Alexander Koller" ]
Recipe texts are an idiosyncratic form of instructional language that pose unique challenges for automatic understanding. One challenge is that a cooking step in one recipe can be explained in another recipe in different words, at a different level of abstraction, or not at all. Previous work has annotated corresponden...
2021.emnlp-main.554
10.18653/v1/2021.emnlp-main.554
null
null
null
2021.emnlp-main.555
Generating Datasets with Pretrained Language Models
https://aclanthology.org/2021.emnlp-main.555/
[ "Timo Schick", "Hinrich Schütze" ]
To obtain high-quality sentence embeddings from pretrained language models (PLMs), they must either be augmented with additional pretraining objectives or finetuned on a large set of labeled text pairs. While the latter approach typically outperforms the former, it requires great human effort to generate suitable datas...
2021.emnlp-main.555
10.18653/v1/2021.emnlp-main.555
null
2104.07540
title_snapshot
2021.emnlp-main.556
Continuous Entailment Patterns for Lexical Inference in Context
https://aclanthology.org/2021.emnlp-main.556/
[ "Martin Schmitt", "Hinrich Schütze" ]
Combining a pretrained language model (PLM) with textual patterns has been shown to help in both zero- and few-shot settings. For zero-shot performance, it makes sense to design patterns that closely resemble the text seen during self-supervised pretraining because the model has never seen anything else. Supervised tra...
2021.emnlp-main.556
10.18653/v1/2021.emnlp-main.556
null
2109.03695
title_snapshot
2021.emnlp-main.557
Numeracy enhances the Literacy of Language Models
https://aclanthology.org/2021.emnlp-main.557/
[ "Avijit Thawani", "Jay Pujara", "Filip Ilievski" ]
Specialized number representations in NLP have shown improvements on numerical reasoning tasks like arithmetic word problems and masked number prediction. But humans also use numeracy to make better sense of world concepts, e.g., you can seat 5 people in your ‘room’ but not 500. Does a better grasp of numbers improve a...
2021.emnlp-main.557
10.18653/v1/2021.emnlp-main.557
null
null
null
2021.emnlp-main.558
Students Who Study Together Learn Better: On the Importance of Collective Knowledge Distillation for Domain Transfer in Fact Verification
https://aclanthology.org/2021.emnlp-main.558/
[ "Mitch Paul Mithun", "Sandeep Suntwal", "Mihai Surdeanu" ]
While neural networks produce state-of-the- art performance in several NLP tasks, they generally depend heavily on lexicalized information, which transfer poorly between domains. Previous works have proposed delexicalization as a form of knowledge distillation to reduce the dependency on such lexical artifacts. However...
2021.emnlp-main.558
10.18653/v1/2021.emnlp-main.558
null
null
null
2021.emnlp-main.559
MultiEURLEX - A multi-lingual and multi-label legal document classification dataset for zero-shot cross-lingual transfer
https://aclanthology.org/2021.emnlp-main.559/
[ "Ilias Chalkidis", "Manos Fergadiotis", "Ion Androutsopoulos" ]
We introduce MULTI-EURLEX, a new multilingual dataset for topic classification of legal documents. The dataset comprises 65k European Union (EU) laws, officially translated in 23 languages, annotated with multiple labels from the EUROVOC taxonomy. We highlight the effect of temporal concept drift and the importance of ...
2021.emnlp-main.559
10.18653/v1/2021.emnlp-main.559
null
2109.00904
title_snapshot
2021.emnlp-main.560
Joint Passage Ranking for Diverse Multi-Answer Retrieval
https://aclanthology.org/2021.emnlp-main.560/
[ "Sewon Min", "Kenton Lee", "Ming-Wei Chang", "Kristina Toutanova", "Hannaneh Hajishirzi" ]
We study multi-answer retrieval, an under-explored problem that requires retrieving passages to cover multiple distinct answers for a given question. This task requires joint modeling of retrieved passages, as models should not repeatedly retrieve passages containing the same answer at the cost of missing a different v...
2021.emnlp-main.560
10.18653/v1/2021.emnlp-main.560
null
2104.08445
title_snapshot
2021.emnlp-main.561
Generative Context Pair Selection for Multi-hop Question Answering
https://aclanthology.org/2021.emnlp-main.561/
[ "Dheeru Dua", "Cicero Nogueira dos Santos", "Patrick Ng", "Ben Athiwaratkun", "Bing Xiang", "Matt Gardner", "Sameer Singh" ]
Compositional reasoning tasks such as multi-hop question answering require models to learn how to make latent decisions using only weak supervision from the final answer. Crowdsourced datasets gathered for these tasks, however, often contain only a slice of the underlying task distribution, which can induce unanticipat...
2021.emnlp-main.561
10.18653/v1/2021.emnlp-main.561
null
2104.08744
title_snapshot
2021.emnlp-main.562
Synthetic Data Augmentation for Zero-Shot Cross-Lingual Question Answering
https://aclanthology.org/2021.emnlp-main.562/
[ "Arij Riabi", "Thomas Scialom", "Rachel Keraron", "Benoît Sagot", "Djamé Seddah", "Jacopo Staiano" ]
Coupled with the availability of large scale datasets, deep learning architectures have enabled rapid progress on the Question Answering task. However, most of those datasets are in English, and the performances of state-of-the-art multilingual models are significantly lower when evaluated on non-English data. Due to h...
2021.emnlp-main.562
10.18653/v1/2021.emnlp-main.562
null
2010.12643
title_snapshot
2021.emnlp-main.563
Have You Seen That Number? Investigating Extrapolation in Question Answering Models
https://aclanthology.org/2021.emnlp-main.563/
[ "Jeonghwan Kim", "Giwon Hong", "Kyung-min Kim", "Junmo Kang", "Sung-Hyon Myaeng" ]
Numerical reasoning in machine reading comprehension (MRC) has shown drastic improvements over the past few years. While the previous models for numerical MRC are able to interpolate the learned numerical reasoning capabilities, it is not clear whether they can perform just as well on numbers unseen in the training dat...
2021.emnlp-main.563
10.18653/v1/2021.emnlp-main.563
null
null
null
2021.emnlp-main.564
Surface Form Competition: Why the Highest Probability Answer Isn’t Always Right
https://aclanthology.org/2021.emnlp-main.564/
[ "Ari Holtzman", "Peter West", "Vered Shwartz", "Yejin Choi", "Luke Zettlemoyer" ]
Large language models have shown promising results in zero-shot settings. For example, they can perform multiple choice tasks simply by conditioning on a question and selecting the answer with the highest probability. However, ranking by string probability can be problematic due to surface form competition—wherein diff...
2021.emnlp-main.564
10.18653/v1/2021.emnlp-main.564
null
2104.08315
title_snapshot
2021.emnlp-main.565
Entity-Based Knowledge Conflicts in Question Answering
https://aclanthology.org/2021.emnlp-main.565/
[ "Shayne Longpre", "Kartik Perisetla", "Anthony Chen", "Nikhil Ramesh", "Chris DuBois", "Sameer Singh" ]
Knowledge-dependent tasks typically use two sources of knowledge: parametric, learned at training time, and contextual, given as a passage at inference time. To understand how models use these sources together, we formalize the problem of knowledge conflicts, where the contextual information contradicts the learned inf...
2021.emnlp-main.565
10.18653/v1/2021.emnlp-main.565
null
2109.05052
title_snapshot
2021.emnlp-main.566
Back-Training excels Self-Training at Unsupervised Domain Adaptation of Question Generation and Passage Retrieval
https://aclanthology.org/2021.emnlp-main.566/
[ "Devang Kulshreshtha", "Robert Belfer", "Iulian Vlad Serban", "Siva Reddy" ]
In this work, we introduce back-training, an alternative to self-training for unsupervised domain adaptation (UDA). While self-training generates synthetic training data where natural inputs are aligned with noisy outputs, back-training results in natural outputs aligned with noisy inputs. This significantly reduces th...
2021.emnlp-main.566
10.18653/v1/2021.emnlp-main.566
null
2104.08801
title_snapshot
2021.emnlp-main.567
DWUG: A large Resource of Diachronic Word Usage Graphs in Four Languages
https://aclanthology.org/2021.emnlp-main.567/
[ "Dominik Schlechtweg", "Nina Tahmasebi", "Simon Hengchen", "Haim Dubossarsky", "Barbara McGillivray" ]
Word meaning is notoriously difficult to capture, both synchronically and diachronically. In this paper, we describe the creation of the largest resource of graded contextualized, diachronic word meaning annotation in four different languages, based on 100,000 human semantic proximity judgments. We describe in detail t...
2021.emnlp-main.567
10.18653/v1/2021.emnlp-main.567
null
2104.08540
title_snapshot
2021.emnlp-main.568
I Wish I Would Have Loved This One, But I Didn’t – A Multilingual Dataset for Counterfactual Detection in Product Review
https://aclanthology.org/2021.emnlp-main.568/
[ "James O’Neill", "Polina Rozenshtein", "Ryuichi Kiryo", "Motoko Kubota", "Danushka Bollegala" ]
Counterfactual statements describe events that did not or cannot take place. We consider the problem of counterfactual detection (CFD) in product reviews. For this purpose, we annotate a multilingual CFD dataset from Amazon product reviews covering counterfactual statements written in English, German, and Japanese lang...
2021.emnlp-main.568
10.18653/v1/2021.emnlp-main.568
null
2104.06893
title_judge
2021.emnlp-main.569
Does It Capture STEL? A Modular, Similarity-based Linguistic Style Evaluation Framework
https://aclanthology.org/2021.emnlp-main.569/
[ "Anna Wegmann", "Dong Nguyen" ]
Style is an integral part of natural language. However, evaluation methods for style measures are rare, often task-specific and usually do not control for content. We propose the modular, fine-grained and content-controlled similarity-based STyle EvaLuation framework (STEL) to test the performance of any model that can...
2021.emnlp-main.569
10.18653/v1/2021.emnlp-main.569
null
2109.04817
title_snapshot
2021.emnlp-main.570
Evaluating the Morphosyntactic Well-formedness of Generated Texts
https://aclanthology.org/2021.emnlp-main.570/
[ "Adithya Pratapa", "Antonios Anastasopoulos", "Shruti Rijhwani", "Aditi Chaudhary", "David R. Mortensen", "Graham Neubig", "Yulia Tsvetkov" ]
Text generation systems are ubiquitous in natural language processing applications. However, evaluation of these systems remains a challenge, especially in multilingual settings. In this paper, we propose L’AMBRE – a metric to evaluate the morphosyntactic well-formedness of text using its dependency parse and morphosyn...
2021.emnlp-main.570
10.18653/v1/2021.emnlp-main.570
null
2103.16590
title_snapshot
2021.emnlp-main.571
AM2iCo: Evaluating Word Meaning in Context across Low-Resource Languages with Adversarial Examples
https://aclanthology.org/2021.emnlp-main.571/
[ "Qianchu Liu", "Edoardo Maria Ponti", "Diana McCarthy", "Ivan Vulić", "Anna Korhonen" ]
Capturing word meaning in context and distinguishing between correspondences and variations across languages is key to building successful multilingual and cross-lingual text representation models. However, existing multilingual evaluation datasets that evaluate lexical semantics “in-context” have various limitations. ...
2021.emnlp-main.571
10.18653/v1/2021.emnlp-main.571
null
2104.08639
title_snapshot
2021.emnlp-main.572
CrossFit: A Few-shot Learning Challenge for Cross-task Generalization in NLP
https://aclanthology.org/2021.emnlp-main.572/
[ "Qinyuan Ye", "Bill Yuchen Lin", "Xiang Ren" ]
Humans can learn a new language task efficiently with only few examples, by leveraging their knowledge obtained when learning prior tasks. In this paper, we explore whether and how such cross-task generalization ability can be acquired, and further applied to build better few-shot learners across diverse NLP tasks. We ...
2021.emnlp-main.572
10.18653/v1/2021.emnlp-main.572
null
2104.08835
title_snapshot
2021.emnlp-main.573
On the Influence of Masking Policies in Intermediate Pre-training
https://aclanthology.org/2021.emnlp-main.573/
[ "Qinyuan Ye", "Belinda Z. Li", "Sinong Wang", "Benjamin Bolte", "Hao Ma", "Wen-tau Yih", "Xiang Ren", "Madian Khabsa" ]
Current NLP models are predominantly trained through a two-stage “pre-train then fine-tune” pipeline. Prior work has shown that inserting an intermediate pre-training stage, using heuristic masking policies for masked language modeling (MLM), can significantly improve final performance. However, it is still unclear (1)...
2021.emnlp-main.573
10.18653/v1/2021.emnlp-main.573
null
2104.08840
title_snapshot
2021.emnlp-main.574
ValNorm Quantifies Semantics to Reveal Consistent Valence Biases Across Languages and Over Centuries
https://aclanthology.org/2021.emnlp-main.574/
[ "Autumn Toney", "Aylin Caliskan" ]
Word embeddings learn implicit biases from linguistic regularities captured by word co-occurrence statistics. By extending methods that quantify human-like biases in word embeddings, we introduce ValNorm, a novel intrinsic evaluation task and method to quantify the valence dimension of affect in human-rated word sets f...
2021.emnlp-main.574
10.18653/v1/2021.emnlp-main.574
null
2006.03950
title_snapshot
2021.emnlp-main.575
Perturbation CheckLists for Evaluating NLG Evaluation Metrics
https://aclanthology.org/2021.emnlp-main.575/
[ "Ananya B. Sai", "Tanay Dixit", "Dev Yashpal Sheth", "Sreyas Mohan", "Mitesh M. Khapra" ]
Natural Language Generation (NLG) evaluation is a multifaceted task requiring assessment of multiple desirable criteria, e.g., fluency, coherency, coverage, relevance, adequacy, overall quality, etc. Across existing datasets for 6 NLG tasks, we observe that the human evaluation scores on these multiple criteria are oft...
2021.emnlp-main.575
10.18653/v1/2021.emnlp-main.575
null
2109.05771
title_snapshot
2021.emnlp-main.576
Robust Open-Vocabulary Translation from Visual Text Representations
https://aclanthology.org/2021.emnlp-main.576/
[ "Elizabeth Salesky", "David Etter", "Matt Post" ]
Machine translation models have discrete vocabularies and commonly use subword segmentation techniques to achieve an ‘open vocabulary.’ This approach relies on consistent and correct underlying unicode sequences, and makes models susceptible to degradation from common types of noise and variation. Motivated by the robu...
2021.emnlp-main.576
10.18653/v1/2021.emnlp-main.576
null
2104.08211
title_snapshot
2021.emnlp-main.577
Don’t Go Far Off: An Empirical Study on Neural Poetry Translation
https://aclanthology.org/2021.emnlp-main.577/
[ "Tuhin Chakrabarty", "Arkadiy Saakyan", "Smaranda Muresan" ]
Despite constant improvements in machine translation quality, automatic poetry translation remains a challenging problem due to the lack of open-sourced parallel poetic corpora, and to the intrinsic complexities involved in preserving the semantics, style and figurative nature of poetry. We present an empirical investi...
2021.emnlp-main.577
10.18653/v1/2021.emnlp-main.577
null
2109.02972
title_snapshot
2021.emnlp-main.578
Improving Multilingual Translation by Representation and Gradient Regularization
https://aclanthology.org/2021.emnlp-main.578/
[ "Yilin Yang", "Akiko Eriguchi", "Alexandre Muzio", "Prasad Tadepalli", "Stefan Lee", "Hany Hassan" ]
Multilingual Neural Machine Translation (NMT) enables one model to serve all translation directions, including ones that are unseen during training, i.e. zero-shot translation. Despite being theoretically attractive, current models often produce low quality translations – commonly failing to even produce outputs in the...
2021.emnlp-main.578
10.18653/v1/2021.emnlp-main.578
null
2109.04778
title_snapshot
2021.emnlp-main.579
Learning Kernel-Smoothed Machine Translation with Retrieved Examples
https://aclanthology.org/2021.emnlp-main.579/
[ "Qingnan Jiang", "Mingxuan Wang", "Jun Cao", "Shanbo Cheng", "Shujian Huang", "Lei Li" ]
How to effectively adapt neural machine translation (NMT) models according to emerging cases without retraining? Despite the great success of neural machine translation, updating the deployed models online remains a challenge. Existing non-parametric approaches that retrieve similar examples from a database to guide th...
2021.emnlp-main.579
10.18653/v1/2021.emnlp-main.579
null
2109.09991
title_snapshot
2021.emnlp-main.580
Uncertainty-Aware Balancing for Multilingual and Multi-Domain Neural Machine Translation Training
https://aclanthology.org/2021.emnlp-main.580/
[ "Minghao Wu", "Yitong Li", "Meng Zhang", "Liangyou Li", "Gholamreza Haffari", "Qun Liu" ]
Learning multilingual and multi-domain translation model is challenging as the heterogeneous and imbalanced data make the model converge inconsistently over different corpora in real world. One common practice is to adjust the share of each corpus in the training, so that the learning process is balanced and low-resour...
2021.emnlp-main.580
10.18653/v1/2021.emnlp-main.580
null
2109.02284
title_snapshot
2021.emnlp-main.581
Universal Simultaneous Machine Translation with Mixture-of-Experts Wait-k Policy
https://aclanthology.org/2021.emnlp-main.581/
[ "Shaolei Zhang", "Yang Feng" ]
Simultaneous machine translation (SiMT) generates translation before reading the entire source sentence and hence it has to trade off between translation quality and latency. To fulfill the requirements of different translation quality and latency in practical applications, the previous methods usually need to train mu...
2021.emnlp-main.581
10.18653/v1/2021.emnlp-main.581
null
2109.05238
title_snapshot
2021.emnlp-main.582
How much coffee was consumed during EMNLP 2019? Fermi Problems: A New Reasoning Challenge for AI
https://aclanthology.org/2021.emnlp-main.582/
[ "Ashwin Kalyan", "Abhinav Kumar", "Arjun Chandrasekaran", "Ashish Sabharwal", "Peter Clark" ]
Many real-world problems require the combined application of multiple reasoning abilities—employing suitable abstractions, commonsense knowledge, and creative synthesis of problem-solving strategies. To help advance AI systems towards such capabilities, we propose a new reasoning challenge, namely Fermi Problems (FPs),...
2021.emnlp-main.582
10.18653/v1/2021.emnlp-main.582
null
2110.14207
title_snapshot
2021.emnlp-main.583
Will this Question be Answered? Question Filtering via Answer Model Distillation for Efficient Question Answering
https://aclanthology.org/2021.emnlp-main.583/
[ "Siddhant Garg", "Alessandro Moschitti" ]
In this paper we propose a novel approach towards improving the efficiency of Question Answering (QA) systems by filtering out questions that will not be answered by them. This is based on an interesting new finding: the answer confidence scores of state-of-the-art QA systems can be approximated well by models solely u...
2021.emnlp-main.583
10.18653/v1/2021.emnlp-main.583
null
2109.07009
title_snapshot
2021.emnlp-main.584
Learning with Instance Bundles for Reading Comprehension
https://aclanthology.org/2021.emnlp-main.584/
[ "Dheeru Dua", "Pradeep Dasigi", "Sameer Singh", "Matt Gardner" ]
When training most modern reading comprehension models, all the questions associated with a context are treated as being independent from each other. However, closely related questions and their corresponding answers are not independent, and leveraging these relationships could provide a strong supervision signal to a ...
2021.emnlp-main.584
10.18653/v1/2021.emnlp-main.584
null
2104.08735
title_snapshot
2021.emnlp-main.585
Explaining Answers with Entailment Trees
https://aclanthology.org/2021.emnlp-main.585/
[ "Bhavana Dalvi", "Peter Jansen", "Oyvind Tafjord", "Zhengnan Xie", "Hannah Smith", "Leighanna Pipatanangkura", "Peter Clark" ]
Our goal, in the context of open-domain textual question-answering (QA), is to explain answers by showing the line of reasoning from what is known to the answer, rather than simply showing a fragment of textual evidence (a “rationale”). If this could be done, new opportunities for understanding and debugging the system...
2021.emnlp-main.585
10.18653/v1/2021.emnlp-main.585
null
2104.08661
title_snapshot
2021.emnlp-main.586
SituatedQA: Incorporating Extra-Linguistic Contexts into QA
https://aclanthology.org/2021.emnlp-main.586/
[ "Michael Zhang", "Eunsol Choi" ]
Answers to the same question may change depending on the extra-linguistic contexts (when and where the question was asked). To study this challenge, we introduce SituatedQA, an open-retrieval QA dataset where systems must produce the correct answer to a question given the temporal or geographical context. To construct ...
2021.emnlp-main.586
10.18653/v1/2021.emnlp-main.586
null
2109.06157
title_snapshot
2021.emnlp-main.587
ConvAbuse: Data, Analysis, and Benchmarks for Nuanced Abuse Detection in Conversational AI
https://aclanthology.org/2021.emnlp-main.587/
[ "Amanda Cercas Curry", "Gavin Abercrombie", "Verena Rieser" ]
We present the first English corpus study on abusive language towards three conversational AI systems gathered ‘in the wild’: an open-domain social bot, a rule-based chatbot, and a task-based system. To account for the complexity of the task, we take a more ‘nuanced’ approach where our ConvAI dataset reflects fine-grai...
2021.emnlp-main.587
10.18653/v1/2021.emnlp-main.587
null
2109.09483
title_snapshot
2021.emnlp-main.588
Conversational Multi-Hop Reasoning with Neural Commonsense Knowledge and Symbolic Logic Rules
https://aclanthology.org/2021.emnlp-main.588/
[ "Forough Arabshahi", "Jennifer Lee", "Antoine Bosselut", "Yejin Choi", "Tom Mitchell" ]
One of the challenges faced by conversational agents is their inability to identify unstated presumptions of their users’ commands, a task trivial for humans due to their common sense. In this paper, we propose a zero-shot commonsense reasoning system for conversational agents in an attempt to achieve this. Our reasone...
2021.emnlp-main.588
10.18653/v1/2021.emnlp-main.588
null
2109.08544
title_snapshot
2021.emnlp-main.589
Towards Automatic Evaluation of Dialog Systems: A Model-Free Off-Policy Evaluation Approach
https://aclanthology.org/2021.emnlp-main.589/
[ "Haoming Jiang", "Bo Dai", "Mengjiao Yang", "Tuo Zhao", "Wei Wei" ]
Reliable automatic evaluation of dialogue systems under an interactive environment has long been overdue. An ideal environment for evaluating dialog systems, also known as the Turing test, needs to involve human interaction, which is usually not affordable for large-scale experiments. Though researchers have attempted ...
2021.emnlp-main.589
10.18653/v1/2021.emnlp-main.589
null
2102.10242
title_snapshot
2021.emnlp-main.590
Continual Learning in Task-Oriented Dialogue Systems
https://aclanthology.org/2021.emnlp-main.590/
[ "Andrea Madotto", "Zhaojiang Lin", "Zhenpeng Zhou", "Seungwhan Moon", "Paul Crook", "Bing Liu", "Zhou Yu", "Eunjoon Cho", "Pascale Fung", "Zhiguang Wang" ]
Continual learning in task-oriented dialogue systems allows the system to add new domains and functionalities overtime after deployment, without incurring the high cost of retraining the whole system each time. In this paper, we propose a first-ever continual learning benchmark for task-oriented dialogue systems with 3...
2021.emnlp-main.590
10.18653/v1/2021.emnlp-main.590
null
2012.15504
title_snapshot
2021.emnlp-main.591
Multilingual and Cross-Lingual Intent Detection from Spoken Data
https://aclanthology.org/2021.emnlp-main.591/
[ "Daniela Gerz", "Pei-Hao Su", "Razvan Kusztos", "Avishek Mondal", "Michał Lis", "Eshan Singhal", "Nikola Mrkšić", "Tsung-Hsien Wen", "Ivan Vulić" ]
We present a systematic study on multilingual and cross-lingual intent detection (ID) from spoken data. The study leverages a new resource put forth in this work, termed MInDS-14, a first training and evaluation resource for the ID task with spoken data. It covers 14 intents extracted from a commercial system in the e-...
2021.emnlp-main.591
10.18653/v1/2021.emnlp-main.591
null
2104.08524
title_snapshot
2021.emnlp-main.592
Investigating Robustness of Dialog Models to Popular Figurative Language Constructs
https://aclanthology.org/2021.emnlp-main.592/
[ "Harsh Jhamtani", "Varun Gangal", "Eduard Hovy", "Taylor Berg-Kirkpatrick" ]
Humans often employ figurative language use in communication, including during interactions with dialog systems. Thus, it is important for real-world dialog systems to be able to handle popular figurative language constructs like metaphor and simile. In this work, we analyze the performance of existing dialog models in...
2021.emnlp-main.592
10.18653/v1/2021.emnlp-main.592
null
2110.00687
title_snapshot
2021.emnlp-main.593
Effective Sequence-to-Sequence Dialogue State Tracking
https://aclanthology.org/2021.emnlp-main.593/
[ "Jeffrey Zhao", "Mahdis Mahdieh", "Ye Zhang", "Yuan Cao", "Yonghui Wu" ]
Sequence-to-sequence models have been applied to a wide variety of NLP tasks, but how to properly use them for dialogue state tracking has not been systematically investigated. In this paper, we study this problem from the perspectives of pre-training objectives as well as the formats of context representations. We dem...
2021.emnlp-main.593
10.18653/v1/2021.emnlp-main.593
null
2108.13990
title_snapshot
2021.emnlp-main.594
MSˆ2: Multi-Document Summarization of Medical Studies
https://aclanthology.org/2021.emnlp-main.594/
[ "Jay DeYoung", "Iz Beltagy", "Madeleine van Zuylen", "Bailey Kuehl", "Lucy Lu Wang" ]
To assess the effectiveness of any medical intervention, researchers must conduct a time-intensive and manual literature review. NLP systems can help to automate or assist in parts of this expensive process. In support of this goal, we release MSˆ2 (Multi-Document Summarization of Medical Studies), a dataset of over 47...
2021.emnlp-main.594
10.18653/v1/2021.emnlp-main.594
null
2104.06486
title_judge
2021.emnlp-main.595
CLIPScore: A Reference-free Evaluation Metric for Image Captioning
https://aclanthology.org/2021.emnlp-main.595/
[ "Jack Hessel", "Ari Holtzman", "Maxwell Forbes", "Ronan Le Bras", "Yejin Choi" ]
Image captioning has conventionally relied on reference-based automatic evaluations, where machine captions are compared against captions written by humans. This is in contrast to the reference-free manner in which humans assess caption quality. In this paper, we report the surprising empirical finding that CLIP (Radfo...
2021.emnlp-main.595
10.18653/v1/2021.emnlp-main.595
null
2104.08718
title_snapshot
2021.emnlp-main.596
On the Challenges of Evaluating Compositional Explanations in Multi-Hop Inference: Relevance, Completeness, and Expert Ratings
https://aclanthology.org/2021.emnlp-main.596/
[ "Peter Jansen", "Kelly J. Smith", "Dan Moreno", "Huitzilin Ortiz" ]
Building compositional explanations requires models to combine two or more facts that, together, describe why the answer to a question is correct. Typically, these “multi-hop” explanations are evaluated relative to one (or a small number of) gold explanations. In this work, we show these evaluations substantially under...
2021.emnlp-main.596
10.18653/v1/2021.emnlp-main.596
null
2109.03334
title_snapshot
2021.emnlp-main.597
ESTER: A Machine Reading Comprehension Dataset for Reasoning about Event Semantic Relations
https://aclanthology.org/2021.emnlp-main.597/
[ "Rujun Han", "I-Hung Hsu", "Jiao Sun", "Julia Baylon", "Qiang Ning", "Dan Roth", "Nanyun Peng" ]
Understanding how events are semantically related to each other is the essence of reading comprehension. Recent event-centric reading comprehension datasets focus mostly on event arguments or temporal relations. While these tasks partially evaluate machines’ ability of narrative understanding, human-like reading compre...
2021.emnlp-main.597
10.18653/v1/2021.emnlp-main.597
null
2104.08350
title_judge
2021.emnlp-main.598
RICA: Evaluating Robust Inference Capabilities Based on Commonsense Axioms
https://aclanthology.org/2021.emnlp-main.598/
[ "Pei Zhou", "Rahul Khanna", "Seyeon Lee", "Bill Yuchen Lin", "Daniel Ho", "Jay Pujara", "Xiang Ren" ]
Pre-trained language models (PTLMs) have achieved impressive performance on commonsense inference benchmarks, but their ability to employ commonsense to make robust inferences, which is crucial for effective communications with humans, is debated. In the pursuit of advancing fluid human-AI communication, we propose a n...
2021.emnlp-main.598
10.18653/v1/2021.emnlp-main.598
null
2005.00782
title_snapshot
2021.emnlp-main.599
Compression, Transduction, and Creation: A Unified Framework for Evaluating Natural Language Generation
https://aclanthology.org/2021.emnlp-main.599/
[ "Mingkai Deng", "Bowen Tan", "Zhengzhong Liu", "Eric Xing", "Zhiting Hu" ]
Natural language generation (NLG) spans a broad range of tasks, each of which serves for specific objectives and desires different properties of generated text. The complexity makes automatic evaluation of NLG particularly challenging. Previous work has typically focused on a single task and developed individual evalua...
2021.emnlp-main.599
10.18653/v1/2021.emnlp-main.599
null
2109.06379
title_snapshot
2021.emnlp-main.600
MATE: Multi-view Attention for Table Transformer Efficiency
https://aclanthology.org/2021.emnlp-main.600/
[ "Julian Eisenschlos", "Maharshi Gor", "Thomas Müller", "William Cohen" ]
This work presents a sparse-attention Transformer architecture for modeling documents that contain large tables. Tables are ubiquitous on the web, and are rich in information. However, more than 20% of relational tables on the web have 20 or more rows (Cafarella et al., 2008), and these large tables present a challenge...
2021.emnlp-main.600
10.18653/v1/2021.emnlp-main.600
null
2109.04312
title_snapshot