paper_id
stringlengths
17
19
title
stringlengths
23
140
paper_url
stringlengths
43
45
authors
listlengths
1
24
abstract
large_stringlengths
262
1.75k
anthology_id
stringlengths
17
19
doi
stringlengths
29
31
award
stringclasses
5 values
arxiv_id
stringlengths
10
10
arxiv_id_source
stringclasses
2 values
2021.naacl-main.201
A Survey on Recent Approaches for Natural Language Processing in Low-Resource Scenarios
https://aclanthology.org/2021.naacl-main.201/
[ "Michael A. Hedderich", "Lukas Lange", "Heike Adel", "Jannik Strötgen", "Dietrich Klakow" ]
Deep neural networks and huge language models are becoming omnipresent in natural language applications. As they are known for requiring large amounts of training data, there is a growing body of work to improve the performance in low-resource settings. Motivated by the recent fundamental changes towards neural models ...
2021.naacl-main.201
10.18653/v1/2021.naacl-main.201
null
2010.12309
title_snapshot
2021.naacl-main.202
Temporal Knowledge Graph Completion using a Linear Temporal Regularizer and Multivector Embeddings
https://aclanthology.org/2021.naacl-main.202/
[ "Chengjin Xu", "Yung-Yu Chen", "Mojtaba Nayyeri", "Jens Lehmann" ]
Representation learning approaches for knowledge graphs have been mostly designed for static data. However, many knowledge graphs involve evolving data, e.g., the fact (The President of the United States is Barack Obama) is valid only from 2009 to 2017. This introduces important challenges for knowledge representation ...
2021.naacl-main.202
10.18653/v1/2021.naacl-main.202
null
null
null
2021.naacl-main.203
UDALM: Unsupervised Domain Adaptation through Language Modeling
https://aclanthology.org/2021.naacl-main.203/
[ "Constantinos Karouzos", "Georgios Paraskevopoulos", "Alexandros Potamianos" ]
In this work we explore Unsupervised Domain Adaptation (UDA) of pretrained language models for downstream tasks. We introduce UDALM, a fine-tuning procedure, using a mixed classification and Masked Language Model loss, that can adapt to the target domain distribution in a robust and sample efficient manner. Our experim...
2021.naacl-main.203
10.18653/v1/2021.naacl-main.203
null
2104.07078
title_snapshot
2021.naacl-main.204
Beyond Black & White: Leveraging Annotator Disagreement via Soft-Label Multi-Task Learning
https://aclanthology.org/2021.naacl-main.204/
[ "Tommaso Fornaciari", "Alexandra Uma", "Silviu Paun", "Barbara Plank", "Dirk Hovy", "Massimo Poesio" ]
Supervised learning assumes that a ground truth label exists. However, the reliability of this ground truth depends on human annotators, who often disagree. Prior work has shown that this disagreement can be helpful in training models. We propose a novel method to incorporate this disagreement as information: in additi...
2021.naacl-main.204
10.18653/v1/2021.naacl-main.204
null
null
null
2021.naacl-main.205
Clustering-based Inference for Biomedical Entity Linking
https://aclanthology.org/2021.naacl-main.205/
[ "Rico Angell", "Nicholas Monath", "Sunil Mohan", "Nishant Yadav", "Andrew McCallum" ]
Due to large number of entities in biomedical knowledge bases, only a small fraction of entities have corresponding labelled training data. This necessitates entity linking models which are able to link mentions of unseen entities using learned representations of entities. Previous approaches link each mention independ...
2021.naacl-main.205
10.18653/v1/2021.naacl-main.205
null
2010.11253
title_snapshot
2021.naacl-main.206
Variance-reduced First-order Meta-learning for Natural Language Processing Tasks
https://aclanthology.org/2021.naacl-main.206/
[ "Lingxiao Wang", "Kevin Huang", "Tengyu Ma", "Quanquan Gu", "Jing Huang" ]
First-order meta-learning algorithms have been widely used in practice to learn initial model parameters that can be quickly adapted to new tasks due to their efficiency and effectiveness. However, existing studies find that meta-learner can overfit to some specific adaptation when we have heterogeneous tasks, leading ...
2021.naacl-main.206
10.18653/v1/2021.naacl-main.206
null
null
null
2021.naacl-main.207
Diversity-Aware Batch Active Learning for Dependency Parsing
https://aclanthology.org/2021.naacl-main.207/
[ "Tianze Shi", "Adrian Benton", "Igor Malioutov", "Ozan İrsoy" ]
While the predictive performance of modern statistical dependency parsers relies heavily on the availability of expensive expert-annotated treebank data, not all annotations contribute equally to the training of the parsers. In this paper, we attempt to reduce the number of labeled examples needed to train a strong dep...
2021.naacl-main.207
10.18653/v1/2021.naacl-main.207
null
2104.13936
title_snapshot
2021.naacl-main.208
How many data points is a prompt worth?
https://aclanthology.org/2021.naacl-main.208/
[ "Teven Le Scao", "Alexander Rush" ]
When fine-tuning pretrained models for classification, researchers either use a generic model head or a task-specific prompt for prediction. Proponents of prompting have argued that prompts provide a method for injecting task-specific guidance, which is beneficial in low-data regimes. We aim to quantify this benefit th...
2021.naacl-main.208
10.18653/v1/2021.naacl-main.208
Outstanding Short Paper
2103.08493
title_snapshot
2021.naacl-main.209
Can Latent Alignments Improve Autoregressive Machine Translation?
https://aclanthology.org/2021.naacl-main.209/
[ "Adi Haviv", "Lior Vassertail", "Omer Levy" ]
Latent alignment objectives such as CTC and AXE significantly improve non-autoregressive machine translation models. Can they improve autoregressive models as well? We explore the possibility of training autoregressive machine translation models with latent alignment objectives, and observe that, in practice, this appr...
2021.naacl-main.209
10.18653/v1/2021.naacl-main.209
null
2104.09554
title_snapshot
2021.naacl-main.210
Smoothing and Shrinking the Sparse Seq2Seq Search Space
https://aclanthology.org/2021.naacl-main.210/
[ "Ben Peters", "André F. T. Martins" ]
Current sequence-to-sequence models are trained to minimize cross-entropy and use softmax to compute the locally normalized probabilities over target sequences. While this setup has led to strong results in a variety of tasks, one unsatisfying aspect is its length bias: models give high scores to short, inadequate hypo...
2021.naacl-main.210
10.18653/v1/2021.naacl-main.210
null
2103.10291
title_snapshot
2021.naacl-main.211
Unified Pre-training for Program Understanding and Generation
https://aclanthology.org/2021.naacl-main.211/
[ "Wasi Ahmad", "Saikat Chakraborty", "Baishakhi Ray", "Kai-Wei Chang" ]
Code summarization and generation empower conversion between programming language (PL) and natural language (NL), while code translation avails the migration of legacy code from one PL to another. This paper introduces PLBART, a sequence-to-sequence model capable of performing a broad spectrum of program and language u...
2021.naacl-main.211
10.18653/v1/2021.naacl-main.211
null
2103.06333
title_snapshot
2021.naacl-main.212
Hyperparameter-free Continuous Learning for Domain Classification in Natural Language Understanding
https://aclanthology.org/2021.naacl-main.212/
[ "Ting Hua", "Yilin Shen", "Changsheng Zhao", "Yen-Chang Hsu", "Hongxia Jin" ]
Domain classification is the fundamental task in natural language understanding (NLU), which often requires fast accommodation to new emerging domains. This constraint makes it impossible to retrain all previous domains, even if they are accessible to the new model. Most existing continual learning approaches suffer fr...
2021.naacl-main.212
10.18653/v1/2021.naacl-main.212
null
2201.01420
title_snapshot
2021.naacl-main.213
On the Embeddings of Variables in Recurrent Neural Networks for Source Code
https://aclanthology.org/2021.naacl-main.213/
[ "Nadezhda Chirkova" ]
Source code processing heavily relies on the methods widely used in natural language processing (NLP), but involves specifics that need to be taken into account to achieve higher quality. An example of this specificity is that the semantics of a variable is defined not only by its name but also by the contexts in which...
2021.naacl-main.213
10.18653/v1/2021.naacl-main.213
null
2010.12693
title_snapshot
2021.naacl-main.214
Cross-Lingual Word Embedding Refinement by \ell_{1} Norm Optimisation
https://aclanthology.org/2021.naacl-main.214/
[ "Xutan Peng", "Chenghua Lin", "Mark Stevenson" ]
Cross-Lingual Word Embeddings (CLWEs) encode words from two or more languages in a shared high-dimensional space in which vectors representing words with similar meaning (regardless of language) are closely located. Existing methods for building high-quality CLWEs learn mappings that minimise the ℓ2 norm loss function....
2021.naacl-main.214
10.18653/v1/2021.naacl-main.214
null
2104.04916
title_snapshot
2021.naacl-main.215
Semantic Frame Forecast
https://aclanthology.org/2021.naacl-main.215/
[ "Chieh-Yang Huang", "Ting-Hao Huang" ]
This paper introduces Semantic Frame Forecast, a task that predicts the semantic frames that will occur in the next 10, 100, or even 1,000 sentences in a running story. Prior work focused on predicting the immediate future of a story, such as one to a few sentences ahead. However, when novelists write long stories, gen...
2021.naacl-main.215
10.18653/v1/2021.naacl-main.215
null
2104.05604
title_snapshot
2021.naacl-main.216
MUSER: MUltimodal Stress detection using Emotion Recognition as an Auxiliary Task
https://aclanthology.org/2021.naacl-main.216/
[ "Yiqun Yao", "Michalis Papakostas", "Mihai Burzo", "Mohamed Abouelenien", "Rada Mihalcea" ]
The capability to automatically detect human stress can benefit artificial intelligent agents involved in affective computing and human-computer interaction. Stress and emotion are both human affective states, and stress has proven to have important implications on the regulation and expression of emotion. Although a s...
2021.naacl-main.216
10.18653/v1/2021.naacl-main.216
null
2105.08146
title_snapshot
2021.naacl-main.217
Learning to Decompose and Organize Complex Tasks
https://aclanthology.org/2021.naacl-main.217/
[ "Yi Zhang", "Sujay Kumar Jauhar", "Julia Kiseleva", "Ryen White", "Dan Roth" ]
People rely on digital task management tools, such as email or to-do apps, to manage their tasks. Some of these tasks are large and complex, leading to action paralysis and feelings of being overwhelmed on the part of the user. The micro-productivity literature has shown that such tasks could benefit from being decompo...
2021.naacl-main.217
10.18653/v1/2021.naacl-main.217
null
null
null
2021.naacl-main.218
Continual Learning for Text Classification with Information Disentanglement Based Regularization
https://aclanthology.org/2021.naacl-main.218/
[ "Yufan Huang", "Yanzhe Zhang", "Jiaao Chen", "Xuezhi Wang", "Diyi Yang" ]
Continual learning has become increasingly important as it enables NLP models to constantly learn and gain knowledge over time. Previous continual learning methods are mainly designed to preserve knowledge from previous tasks, without much emphasis on how to well generalize models to new tasks. In this work, we propose...
2021.naacl-main.218
10.18653/v1/2021.naacl-main.218
null
2104.05489
title_snapshot
2021.naacl-main.219
Learning from Executions for Semantic Parsing
https://aclanthology.org/2021.naacl-main.219/
[ "Bailin Wang", "Mirella Lapata", "Ivan Titov" ]
Semantic parsing aims at translating natural language (NL) utterances onto machine-interpretable programs, which can be executed against a real-world environment. The expensive annotation of utterance-program pairs has long been acknowledged as a major bottleneck for the deployment of contemporary neural models to real...
2021.naacl-main.219
10.18653/v1/2021.naacl-main.219
null
2104.05819
title_snapshot
2021.naacl-main.220
Learning to Synthesize Data for Semantic Parsing
https://aclanthology.org/2021.naacl-main.220/
[ "Bailin Wang", "Wenpeng Yin", "Xi Victoria Lin", "Caiming Xiong" ]
Synthesizing data for semantic parsing has gained increasing attention recently. However, most methods require handcrafted (high-precision) rules in their generative process, hindering the exploration of diverse unseen data. In this work, we propose a generative model which features a (non-neural) PCFG that models the ...
2021.naacl-main.220
10.18653/v1/2021.naacl-main.220
null
2104.05827
title_snapshot
2021.naacl-main.221
Edge: Enriching Knowledge Graph Embeddings with External Text
https://aclanthology.org/2021.naacl-main.221/
[ "Saed Rezayi", "Handong Zhao", "Sungchul Kim", "Ryan Rossi", "Nedim Lipka", "Sheng Li" ]
Knowledge graphs suffer from sparsity which degrades the quality of representations generated by various methods. While there is an abundance of textual information throughout the web and many existing knowledge bases, aligning information across these diverse data sources remains a challenge in the literature. Previou...
2021.naacl-main.221
10.18653/v1/2021.naacl-main.221
null
2104.04909
title_snapshot
2021.naacl-main.222
FLIN: A Flexible Natural Language Interface for Web Navigation
https://aclanthology.org/2021.naacl-main.222/
[ "Sahisnu Mazumder", "Oriana Riva" ]
AI assistants can now carry out tasks for users by directly interacting with website UIs. Current semantic parsing and slot-filling techniques cannot flexibly adapt to many different websites without being constantly re-trained. We propose FLIN, a natural language interface for web navigation that maps user commands to...
2021.naacl-main.222
10.18653/v1/2021.naacl-main.222
null
2010.12844
title_snapshot
2021.naacl-main.223
Game-theoretic Vocabulary Selection via the Shapley Value and Banzhaf Index
https://aclanthology.org/2021.naacl-main.223/
[ "Roma Patel", "Marta Garnelo", "Ian Gemp", "Chris Dyer", "Yoram Bachrach" ]
The input vocabulary and the representations learned are crucial to the performance of neural NLP models. Using the full vocabulary results in less explainable and more memory intensive models, with the embedding layer often constituting the majority of model parameters. It is thus common to use a smaller vocabulary to...
2021.naacl-main.223
10.18653/v1/2021.naacl-main.223
null
null
null
2021.naacl-main.224
Incorporating External Knowledge to Enhance Tabular Reasoning
https://aclanthology.org/2021.naacl-main.224/
[ "J. Neeraja", "Vivek Gupta", "Vivek Srikumar" ]
Reasoning about tabular information presents unique challenges to modern NLP approaches which largely rely on pre-trained contextualized embeddings of text. In this paper, we study these challenges through the problem of tabular natural language inference. We propose easy and effective modifications to how information ...
2021.naacl-main.224
10.18653/v1/2021.naacl-main.224
null
2104.04243
title_snapshot
2021.naacl-main.225
Compositional Generalization for Neural Semantic Parsing via Span-level Supervised Attention
https://aclanthology.org/2021.naacl-main.225/
[ "Pengcheng Yin", "Hao Fang", "Graham Neubig", "Adam Pauls", "Emmanouil Antonios Platanios", "Yu Su", "Sam Thomson", "Jacob Andreas" ]
We describe a span-level supervised attention loss that improves compositional generalization in semantic parsers. Our approach builds on existing losses that encourage attention maps in neural sequence-to-sequence models to imitate the output of classical word alignment algorithms. Where past work has used word-level ...
2021.naacl-main.225
10.18653/v1/2021.naacl-main.225
null
null
null
2021.naacl-main.226
Domain Adaptation for Arabic Cross-Domain and Cross-Dialect Sentiment Analysis from Contextualized Word Embedding
https://aclanthology.org/2021.naacl-main.226/
[ "Abdellah El Mekki", "Abdelkader El Mahdaouy", "Ismail Berrada", "Ahmed Khoumsi" ]
Finetuning deep pre-trained language models has shown state-of-the-art performances on a wide range of Natural Language Processing (NLP) applications. Nevertheless, their generalization performance drops under domain shift. In the case of Arabic language, diglossia makes building and annotating corpora for each dialect...
2021.naacl-main.226
10.18653/v1/2021.naacl-main.226
null
null
null
2021.naacl-main.227
Multi-task Learning of Negation and Speculation for Targeted Sentiment Classification
https://aclanthology.org/2021.naacl-main.227/
[ "Andrew Moore", "Jeremy Barnes" ]
The majority of work in targeted sentiment analysis has concentrated on finding better methods to improve the overall results. Within this paper we show that these models are not robust to linguistic phenomena, specifically negation and speculation. In this paper, we propose a multi-task learning method to incorporate ...
2021.naacl-main.227
10.18653/v1/2021.naacl-main.227
null
2010.08318
title_snapshot
2021.naacl-main.228
A Disentangled Adversarial Neural Topic Model for Separating Opinions from Plots in User Reviews
https://aclanthology.org/2021.naacl-main.228/
[ "Gabriele Pergola", "Lin Gui", "Yulan He" ]
The flexibility of the inference process in Variational Autoencoders (VAEs) has recently led to revising traditional probabilistic topic models giving rise to Neural Topic Models (NTM). Although these approaches have achieved significant results, surprisingly very little work has been done on how to disentangle the lat...
2021.naacl-main.228
10.18653/v1/2021.naacl-main.228
null
2010.11384
title_snapshot
2021.naacl-main.229
Graph Ensemble Learning over Multiple Dependency Trees for Aspect-level Sentiment Classification
https://aclanthology.org/2021.naacl-main.229/
[ "Xiaochen Hou", "Peng Qi", "Guangtao Wang", "Rex Ying", "Jing Huang", "Xiaodong He", "Bowen Zhou" ]
Recent work on aspect-level sentiment classification has demonstrated the efficacy of incorporating syntactic structures such as dependency trees with graph neural networks (GNN), but these approaches are usually vulnerable to parsing errors. To better leverage syntactic information in the face of unavoidable errors, w...
2021.naacl-main.229
10.18653/v1/2021.naacl-main.229
null
2103.11794
title_snapshot
2021.naacl-main.230
Emotion-Infused Models for Explainable Psychological Stress Detection
https://aclanthology.org/2021.naacl-main.230/
[ "Elsbeth Turcan", "Smaranda Muresan", "Kathleen McKeown" ]
The problem of detecting psychological stress in online posts, and more broadly, of detecting people in distress or in need of help, is a sensitive application for which the ability to interpret models is vital. Here, we present work exploring the use of a semantically related task, emotion detection, for equally compe...
2021.naacl-main.230
10.18653/v1/2021.naacl-main.230
null
null
null
2021.naacl-main.231
Aspect-based Sentiment Analysis with Type-aware Graph Convolutional Networks and Layer Ensemble
https://aclanthology.org/2021.naacl-main.231/
[ "Yuanhe Tian", "Guimin Chen", "Yan Song" ]
It is popular that neural graph-based models are applied in existing aspect-based sentiment analysis (ABSA) studies for utilizing word relations through dependency parses to facilitate the task with better semantic guidance for analyzing context and aspect words. However, most of these studies only leverage dependency ...
2021.naacl-main.231
10.18653/v1/2021.naacl-main.231
null
null
null
2021.naacl-main.232
Supertagging-based Parsing with Linear Context-free Rewriting Systems
https://aclanthology.org/2021.naacl-main.232/
[ "Thomas Ruprecht", "Richard Mörbitz" ]
We present the first supertagging-based parser for linear context-free rewriting systems (LCFRS). It utilizes neural classifiers and outperforms previous LCFRS-based parsers in both accuracy and parsing speed by a wide margin. Our results keep up with the best (general) discontinuous parsers, particularly the scores fo...
2021.naacl-main.232
10.18653/v1/2021.naacl-main.232
null
2010.10238
title_snapshot
2021.naacl-main.233
Outside Computation with Superior Functions
https://aclanthology.org/2021.naacl-main.233/
[ "Parker Riley", "Daniel Gildea" ]
We show that a general algorithm for efficient computation of outside values under the minimum of superior functions framework proposed by Knuth (1977) would yield a sub-exponential time algorithm for SAT, violating the Strong Exponential Time Hypothesis (SETH).
2021.naacl-main.233
10.18653/v1/2021.naacl-main.233
null
null
null
2021.naacl-main.234
Learning Syntax from Naturally-Occurring Bracketings
https://aclanthology.org/2021.naacl-main.234/
[ "Tianze Shi", "Ozan İrsoy", "Igor Malioutov", "Lillian Lee" ]
Naturally-occurring bracketings, such as answer fragments to natural language questions and hyperlinks on webpages, can reflect human syntactic intuition regarding phrasal boundaries. Their availability and approximate correspondence to syntax make them appealing as distant information sources to incorporate into unsup...
2021.naacl-main.234
10.18653/v1/2021.naacl-main.234
null
2104.13933
title_snapshot
2021.naacl-main.235
Bot-Adversarial Dialogue for Safe Conversational Agents
https://aclanthology.org/2021.naacl-main.235/
[ "Jing Xu", "Da Ju", "Margaret Li", "Y-Lan Boureau", "Jason Weston", "Emily Dinan" ]
Conversational agents trained on large unlabeled corpora of human interactions will learn patterns and mimic behaviors therein, which include offensive or otherwise toxic behavior. We introduce a new human-and-model-in-the-loop framework for evaluating the toxicity of such models, and compare a variety of existing meth...
2021.naacl-main.235
10.18653/v1/2021.naacl-main.235
null
null
null
2021.naacl-main.236
Non-Autoregressive Semantic Parsing for Compositional Task-Oriented Dialog
https://aclanthology.org/2021.naacl-main.236/
[ "Arun Babu", "Akshat Shrivastava", "Armen Aghajanyan", "Ahmed Aly", "Angela Fan", "Marjan Ghazvininejad" ]
Semantic parsing using sequence-to-sequence models allows parsing of deeper representations compared to traditional word tagging based models. In spite of these advantages, widespread adoption of these models for real-time conversational use cases has been stymied by higher compute requirements and thus higher latency....
2021.naacl-main.236
10.18653/v1/2021.naacl-main.236
null
2104.04923
title_snapshot
2021.naacl-main.237
Example-Driven Intent Prediction with Observers
https://aclanthology.org/2021.naacl-main.237/
[ "Shikib Mehri", "Mihail Eric" ]
A key challenge of dialog systems research is to effectively and efficiently adapt to new domains. A scalable paradigm for adaptation necessitates the development of generalizable models that perform well in few-shot settings. In this paper, we focus on the intent classification problem which aims to identify user inte...
2021.naacl-main.237
10.18653/v1/2021.naacl-main.237
null
2010.08684
title_snapshot
2021.naacl-main.238
Imperfect also Deserves Reward: Multi-Level and Sequential Reward Modeling for Better Dialog Management
https://aclanthology.org/2021.naacl-main.238/
[ "Zhengxu Hou", "Bang Liu", "Ruihui Zhao", "Zijing Ou", "Yafei Liu", "Xi Chen", "Yefeng Zheng" ]
For task-oriented dialog systems, training a Reinforcement Learning (RL) based Dialog Management module suffers from low sample efficiency and slow convergence speed due to the sparse rewards in RL. To solve this problem, many strategies have been proposed to give proper rewards when training RL, but their rewards lack...
2021.naacl-main.238
10.18653/v1/2021.naacl-main.238
null
2104.04748
title_snapshot
2021.naacl-main.239
Action-Based Conversations Dataset: A Corpus for Building More In-Depth Task-Oriented Dialogue Systems
https://aclanthology.org/2021.naacl-main.239/
[ "Derek Chen", "Howard Chen", "Yi Yang", "Alexander Lin", "Zhou Yu" ]
Existing goal-oriented dialogue datasets focus mainly on identifying slots and values. However, customer support interactions in reality often involve agents following multi-step procedures derived from explicitly-defined company policies as well. To study customer service dialogue systems in more realistic settings, w...
2021.naacl-main.239
10.18653/v1/2021.naacl-main.239
null
2104.00783
title_snapshot
2021.naacl-main.240
Controlling Dialogue Generation with Semantic Exemplars
https://aclanthology.org/2021.naacl-main.240/
[ "Prakhar Gupta", "Jeffrey Bigham", "Yulia Tsvetkov", "Amy Pavel" ]
Dialogue systems pretrained with large language models generate locally coherent responses, but lack fine-grained control over responses necessary to achieve specific goals. A promising method to control response generation is exemplar-based generation, in which models edit exemplar responses that are retrieved from tr...
2021.naacl-main.240
10.18653/v1/2021.naacl-main.240
null
2008.09075
title_snapshot
2021.naacl-main.241
COIL: Revisit Exact Lexical Match in Information Retrieval with Contextualized Inverted List
https://aclanthology.org/2021.naacl-main.241/
[ "Luyu Gao", "Zhuyun Dai", "Jamie Callan" ]
Classical information retrieval systems such as BM25 rely on exact lexical match and can carry out search efficiently with inverted list index. Recent neural IR models shifts towards soft matching all query document terms, but they lose the computation efficiency of exact match systems. This paper presents COIL, a cont...
2021.naacl-main.241
10.18653/v1/2021.naacl-main.241
null
2104.07186
title_snapshot
2021.naacl-main.242
X-Class: Text Classification with Extremely Weak Supervision
https://aclanthology.org/2021.naacl-main.242/
[ "Zihan Wang", "Dheeraj Mekala", "Jingbo Shang" ]
In this paper, we explore text classification with extremely weak supervision, i.e., only relying on the surface text of class names. This is a more challenging setting than the seed-driven weak supervision, which allows a few seed words per class. We opt to attack this problem from a representation learning perspectiv...
2021.naacl-main.242
10.18653/v1/2021.naacl-main.242
null
2010.12794
title_snapshot
2021.naacl-main.243
Fine-tuning Encoders for Improved Monolingual and Zero-shot Polylingual Neural Topic Modeling
https://aclanthology.org/2021.naacl-main.243/
[ "Aaron Mueller", "Mark Dredze" ]
Neural topic models can augment or replace bag-of-words inputs with the learned representations of deep pre-trained transformer-based word prediction models. One added benefit when using representations from multilingual models is that they facilitate zero-shot polylingual topic modeling. However, while it has been wid...
2021.naacl-main.243
10.18653/v1/2021.naacl-main.243
null
2104.05064
title_snapshot
2021.naacl-main.244
Exploring the Relationship Between Algorithm Performance, Vocabulary, and Run-Time in Text Classification
https://aclanthology.org/2021.naacl-main.244/
[ "Wilson Fearn", "Orion Weller", "Kevin Seppi" ]
Text classification is a significant branch of natural language processing, and has many applications including document classification and sentiment analysis. Unsurprisingly, those who do text classification are concerned with the run-time of their algorithms, many of which depend on the size of the corpus’ vocabulary...
2021.naacl-main.244
10.18653/v1/2021.naacl-main.244
null
2104.03848
title_snapshot
2021.naacl-main.245
Faithfully Explainable Recommendation via Neural Logic Reasoning
https://aclanthology.org/2021.naacl-main.245/
[ "Yaxin Zhu", "Yikun Xian", "Zuohui Fu", "Gerard de Melo", "Yongfeng Zhang" ]
Knowledge graphs (KG) have become increasingly important to endow modern recommender systems with the ability to generate traceable reasoning paths to explain the recommendation process. However, prior research rarely considers the faithfulness of the derived explanations to justify the decision-making process. To the ...
2021.naacl-main.245
10.18653/v1/2021.naacl-main.245
null
2104.07869
title_snapshot
2021.naacl-main.246
You Sound Like Someone Who Watches Drama Movies: Towards Predicting Movie Preferences from Conversational Interactions
https://aclanthology.org/2021.naacl-main.246/
[ "Sergey Volokhin", "Joyce Ho", "Oleg Rokhlenko", "Eugene Agichtein" ]
The increasing popularity of voice-based personal assistants provides new opportunities for conversational recommendation. One particularly interesting area is movie recommendation, which can benefit from an open-ended interaction with the user, through a natural conversation. We explore one promising direction for con...
2021.naacl-main.246
10.18653/v1/2021.naacl-main.246
null
null
null
2021.naacl-main.247
Reading and Acting while Blindfolded: The Need for Semantics in Text Game Agents
https://aclanthology.org/2021.naacl-main.247/
[ "Shunyu Yao", "Karthik Narasimhan", "Matthew Hausknecht" ]
Text-based games simulate worlds and interact with players using natural language. Recent work has used them as a testbed for autonomous language-understanding agents, with the motivation being that understanding the meanings of words or semantics is a key component of how humans understand, reason, and act in these wo...
2021.naacl-main.247
10.18653/v1/2021.naacl-main.247
null
2103.13552
title_snapshot
2021.naacl-main.248
SOrT-ing VQA Models : Contrastive Gradient Learning for Improved Consistency
https://aclanthology.org/2021.naacl-main.248/
[ "Sameer Dharur", "Purva Tendulkar", "Dhruv Batra", "Devi Parikh", "Ramprasaath R. Selvaraju" ]
Recent research in Visual Question Answering (VQA) has revealed state-of-the-art models to be inconsistent in their understanding of the world - they answer seemingly difficult questions requiring reasoning correctly but get simpler associated sub-questions wrong. These sub-questions pertain to lower level visual conce...
2021.naacl-main.248
10.18653/v1/2021.naacl-main.248
null
2010.10038
title_snapshot
2021.naacl-main.249
Semi-Supervised Policy Initialization for Playing Games with Language Hints
https://aclanthology.org/2021.naacl-main.249/
[ "Tsu-Jui Fu", "William Yang Wang" ]
Using natural language as a hint can supply an additional reward for playing sparse-reward games. Achieving a goal should involve several different hints, while the given hints are usually incomplete. Those unmentioned latent hints still rely on the sparse reward signal, and make the learning process difficult. In this...
2021.naacl-main.249
10.18653/v1/2021.naacl-main.249
null
null
null
2021.naacl-main.250
Revisiting Document Representations for Large-Scale Zero-Shot Learning
https://aclanthology.org/2021.naacl-main.250/
[ "Jihyung Kil", "Wei-Lun Chao" ]
Zero-shot learning aims to recognize unseen objects using their semantic representations. Most existing works use visual attributes labeled by humans, not suitable for large-scale applications. In this paper, we revisit the use of documents as semantic representations. We argue that documents like Wikipedia pages conta...
2021.naacl-main.250
10.18653/v1/2021.naacl-main.250
null
2104.10355
title_snapshot
2021.naacl-main.251
Negative language transfer in learner English: A new dataset
https://aclanthology.org/2021.naacl-main.251/
[ "Leticia Farias Wanderley", "Nicole Zhao", "Carrie Demmans Epp" ]
Automatic personalized corrective feedback can help language learners from different backgrounds better acquire a new language. This paper introduces a learner English dataset in which learner errors are accompanied by information about possible error sources. This dataset contains manually annotated error causes for l...
2021.naacl-main.251
10.18653/v1/2021.naacl-main.251
null
null
null
2021.naacl-main.252
SentSim: Crosslingual Semantic Evaluation of Machine Translation
https://aclanthology.org/2021.naacl-main.252/
[ "Yurun Song", "Junchen Zhao", "Lucia Specia" ]
Machine translation (MT) is currently evaluated in one of two ways: in a monolingual fashion, by comparison with the system output to one or more human reference translations, or in a trained crosslingual fashion, by building a supervised model to predict quality scores from human-labeled data. In this paper, we propos...
2021.naacl-main.252
10.18653/v1/2021.naacl-main.252
null
null
null
2021.naacl-main.253
Quality Estimation for Image Captions Based on Large-scale Human Evaluations
https://aclanthology.org/2021.naacl-main.253/
[ "Tomer Levinboim", "Ashish V. Thapliyal", "Piyush Sharma", "Radu Soricut" ]
Automatic image captioning has improved significantly over the last few years, but the problem is far from being solved, with state of the art models still often producing low quality captions when used in the wild. In this paper, we focus on the task of Quality Estimation (QE) for image captions, which attempts to mod...
2021.naacl-main.253
10.18653/v1/2021.naacl-main.253
null
1909.03396
title_snapshot
2021.naacl-main.254
CaSiNo: A Corpus of Campsite Negotiation Dialogues for Automatic Negotiation Systems
https://aclanthology.org/2021.naacl-main.254/
[ "Kushal Chawla", "Jaysa Ramirez", "Rene Clever", "Gale Lucas", "Jonathan May", "Jonathan Gratch" ]
Automated systems that negotiate with humans have broad applications in pedagogy and conversational AI. To advance the development of practical negotiation systems, we present CaSiNo: a novel corpus of over a thousand negotiation dialogues in English. Participants take the role of campsite neighbors and negotiate for f...
2021.naacl-main.254
10.18653/v1/2021.naacl-main.254
null
2103.15721
title_snapshot
2021.naacl-main.255
News Headline Grouping as a Challenging NLU Task
https://aclanthology.org/2021.naacl-main.255/
[ "Philippe Laban", "Lucas Bandarkar", "Marti A. Hearst" ]
Recent progress in Natural Language Understanding (NLU) has seen the latest models outperform human performance on many standard tasks. These impressive results have led the community to introspect on dataset limitations, and iterate on more nuanced challenges. In this paper, we introduce the task of HeadLine Grouping ...
2021.naacl-main.255
10.18653/v1/2021.naacl-main.255
null
2105.05391
title_snapshot
2021.naacl-main.256
Olá, Bonjour, Salve! XFORMAL: A Benchmark for Multilingual Formality Style Transfer
https://aclanthology.org/2021.naacl-main.256/
[ "Eleftheria Briakou", "Di Lu", "Ke Zhang", "Joel Tetreault" ]
We take the first step towards multilingual style transfer by creating and releasing XFORMAL, a benchmark of multiple formal reformulations of informal text in Brazilian Portuguese, French, and Italian. Results on XFORMAL suggest that state-of-the-art style transfer approaches perform close to simple baselines, indicat...
2021.naacl-main.256
10.18653/v1/2021.naacl-main.256
null
2104.04108
title_judge
2021.naacl-main.257
Grouping Words with Semantic Diversity
https://aclanthology.org/2021.naacl-main.257/
[ "Karine Chubarian", "Abdul Rafae Khan", "Anastasios Sidiropoulos", "Jia Xu" ]
Deep Learning-based NLP systems can be sensitive to unseen tokens and hard to learn with high-dimensional inputs, which critically hinder learning generalization. We introduce an approach by grouping input words based on their semantic diversity to simplify input language representation with low ambiguity. Since the se...
2021.naacl-main.257
10.18653/v1/2021.naacl-main.257
null
null
null
2021.naacl-main.258
Noise Stability Regularization for Improving BERT Fine-tuning
https://aclanthology.org/2021.naacl-main.258/
[ "Hang Hua", "Xingjian Li", "Dejing Dou", "Chengzhong Xu", "Jiebo Luo" ]
Fine-tuning pre-trained language models suchas BERT has become a common practice dom-inating leaderboards across various NLP tasks. Despite its recent success and wide adoption,this process is unstable when there are onlya small number of training samples available. The brittleness of this process is often reflectedby ...
2021.naacl-main.258
10.18653/v1/2021.naacl-main.258
null
2107.04835
title_snapshot
2021.naacl-main.259
FlowPrior: Learning Expressive Priors for Latent Variable Sentence Models
https://aclanthology.org/2021.naacl-main.259/
[ "Xiaoan Ding", "Kevin Gimpel" ]
Variational autoencoders (VAEs) are widely used for latent variable modeling of text. We focus on variations that learn expressive prior distributions over the latent variable. We find that existing training strategies are not effective for learning rich priors, so we propose adding the importance-sampled log marginal ...
2021.naacl-main.259
10.18653/v1/2021.naacl-main.259
null
null
null
2021.naacl-main.260
HTCInfoMax: A Global Model for Hierarchical Text Classification via Information Maximization
https://aclanthology.org/2021.naacl-main.260/
[ "Zhongfen Deng", "Hao Peng", "Dongxiao He", "Jianxin Li", "Philip Yu" ]
The current state-of-the-art model HiAGM for hierarchical text classification has two limitations. First, it correlates each text sample with all labels in the dataset which contains irrelevant information. Second, it does not consider any statistical constraint on the label representations learned by the structure enc...
2021.naacl-main.260
10.18653/v1/2021.naacl-main.260
null
2104.05220
title_snapshot
2021.naacl-main.261
Knowledge Guided Metric Learning for Few-Shot Text Classification
https://aclanthology.org/2021.naacl-main.261/
[ "Dianbo Sui", "Yubo Chen", "Binjie Mao", "Delai Qiu", "Kang Liu", "Jun Zhao" ]
Humans can distinguish new categories very efficiently with few examples, largely due to the fact that human beings can leverage knowledge obtained from relevant tasks. However, deep learning based text classification model tends to struggle to achieve satisfactory performance when labeled data are scarce. Inspired by ...
2021.naacl-main.261
10.18653/v1/2021.naacl-main.261
null
2004.01907
title_snapshot
2021.naacl-main.262
Ensemble of MRR and NDCG models for Visual Dialog
https://aclanthology.org/2021.naacl-main.262/
[ "Idan Schwartz" ]
Assessing an AI agent that can converse in human language and understand visual content is challenging. Generation metrics, such as BLEU scores favor correct syntax over semantics. Hence a discriminative approach is often used, where an agent ranks a set of candidate options. The mean reciprocal rank (MRR) metric evalu...
2021.naacl-main.262
10.18653/v1/2021.naacl-main.262
null
2104.07511
title_snapshot
2021.naacl-main.263
Supervised Neural Clustering via Latent Structured Output Learning: Application to Question Intents
https://aclanthology.org/2021.naacl-main.263/
[ "Iryna Haponchyk", "Alessandro Moschitti" ]
Previous pre-neural work on structured prediction has produced very effective supervised clustering algorithms using linear classifiers, e.g., structured SVM or perceptron. However, these cannot exploit the representation learning ability of neural networks, which would make supervised clustering even more powerful, i....
2021.naacl-main.263
10.18653/v1/2021.naacl-main.263
null
null
null
2021.naacl-main.264
ConVEx: Data-Efficient and Few-Shot Slot Labeling
https://aclanthology.org/2021.naacl-main.264/
[ "Matthew Henderson", "Ivan Vulić" ]
We propose ConVEx (Conversational Value Extractor), an efficient pretraining and fine-tuning neural approach for slot-labeling dialog tasks. Instead of relying on more general pretraining objectives from prior work (e.g., language modeling, response selection), ConVEx’s pretraining objective, a novel pairwise cloze tas...
2021.naacl-main.264
10.18653/v1/2021.naacl-main.264
null
2010.11791
title_snapshot
2021.naacl-main.265
CREAD: Combined Resolution of Ellipses and Anaphora in Dialogues
https://aclanthology.org/2021.naacl-main.265/
[ "Bo-Hsiang Tseng", "Shruti Bhargava", "Jiarui Lu", "Joel Ruben Antony Moniz", "Dhivya Piraviperumal", "Lin Li", "Hong Yu" ]
Anaphora and ellipses are two common phenomena in dialogues. Without resolving referring expressions and information omission, dialogue systems may fail to generate consistent and coherent responses. Traditionally, anaphora is resolved by coreference resolution and ellipses by query rewrite. In this work, we propose a ...
2021.naacl-main.265
10.18653/v1/2021.naacl-main.265
null
2105.09914
title_snapshot
2021.naacl-main.266
Knowledge-Driven Slot Constraints for Goal-Oriented Dialogue Systems
https://aclanthology.org/2021.naacl-main.266/
[ "Piyawat Lertvittayakumjorn", "Daniele Bonadiman", "Saab Mansour" ]
In goal-oriented dialogue systems, users provide information through slot values to achieve specific goals. Practically, some combinations of slot values can be invalid according to external knowledge. For example, a combination of “cheese pizza” (a menu item) and “oreo cookies” (a topping) from an input utterance “Can...
2021.naacl-main.266
10.18653/v1/2021.naacl-main.266
null
null
null
2021.naacl-main.267
Clipping Loops for Sample-Efficient Dialogue Policy Optimisation
https://aclanthology.org/2021.naacl-main.267/
[ "Yen-Chen Wu", "Carl Edward Rasmussen" ]
Training dialogue agents requires a large number of interactions with users: agents have no idea about which responses are bad among a lengthy dialogue. In this paper, we propose loop-clipping policy optimisation (LCPO) to eliminate useless responses. LCPO consists of two stages: loop clipping and advantage clipping. I...
2021.naacl-main.267
10.18653/v1/2021.naacl-main.267
null
null
null
2021.naacl-main.268
Integrating Lexical Information into Entity Neighbourhood Representations for Relation Prediction
https://aclanthology.org/2021.naacl-main.268/
[ "Ian Wood", "Mark Johnson", "Stephen Wan" ]
Relation prediction informed from a combination of text corpora and curated knowledge bases, combining knowledge graph completion with relation extraction, is a relatively little studied task. A system that can perform this task has the ability to extend an arbitrary set of relational database tables with information e...
2021.naacl-main.268
10.18653/v1/2021.naacl-main.268
null
null
null
2021.naacl-main.269
Noisy-Labeled NER with Confidence Estimation
https://aclanthology.org/2021.naacl-main.269/
[ "Kun Liu", "Yao Fu", "Chuanqi Tan", "Mosha Chen", "Ningyu Zhang", "Songfang Huang", "Sheng Gao" ]
Recent studies in deep learning have shown significant progress in named entity recognition (NER). However, most existing works assume clean data annotation, while real-world scenarios typically involve a large amount of noises from a variety of sources (e.g., pseudo, weak, or distant annotations). This work studies NE...
2021.naacl-main.269
10.18653/v1/2021.naacl-main.269
null
2104.04318
title_snapshot
2021.naacl-main.270
TABBIE: Pretrained Representations of Tabular Data
https://aclanthology.org/2021.naacl-main.270/
[ "Hiroshi Iida", "Dung Thai", "Varun Manjunatha", "Mohit Iyyer" ]
Existing work on tabular representation-learning jointly models tables and associated text using self-supervised objective functions derived from pretrained language models such as BERT. While this joint pretraining improves tasks involving paired tables and text (e.g., answering questions about tables), we show that i...
2021.naacl-main.270
10.18653/v1/2021.naacl-main.270
null
2105.02584
title_snapshot
2021.naacl-main.271
Better Feature Integration for Named Entity Recognition
https://aclanthology.org/2021.naacl-main.271/
[ "Lu Xu", "Zhanming Jie", "Wei Lu", "Lidong Bing" ]
It has been shown that named entity recognition (NER) could benefit from incorporating the long-distance structured information captured by dependency trees. We believe this is because both types of features - the contextual information captured by the linear sequences and the structured information captured by the dep...
2021.naacl-main.271
10.18653/v1/2021.naacl-main.271
null
2104.05316
title_snapshot
2021.naacl-main.272
ZS-BERT: Towards Zero-Shot Relation Extraction with Attribute Representation Learning
https://aclanthology.org/2021.naacl-main.272/
[ "Chih-Yao Chen", "Cheng-Te Li" ]
While relation extraction is an essential task in knowledge acquisition and representation, and new-generated relations are common in the real world, less effort is made to predict unseen relations that cannot be observed at the training stage. In this paper, we formulate the zero-shot relation extraction problem by in...
2021.naacl-main.272
10.18653/v1/2021.naacl-main.272
null
2104.04697
title_snapshot
2021.naacl-main.273
Graph Convolutional Networks for Event Causality Identification with Rich Document-level Structures
https://aclanthology.org/2021.naacl-main.273/
[ "Minh Tran Phu", "Thien Huu Nguyen" ]
We study the problem of Event Causality Identification (ECI) to detect causal relation between event mention pairs in text. Although deep learning models have recently shown state-of-the-art performance for ECI, they are limited to the intra-sentence setting where event mention pairs are presented in the same sentences...
2021.naacl-main.273
10.18653/v1/2021.naacl-main.273
null
null
null
2021.naacl-main.274
A Context-Dependent Gated Module for Incorporating Symbolic Semantics into Event Coreference Resolution
https://aclanthology.org/2021.naacl-main.274/
[ "Tuan Lai", "Heng Ji", "Trung Bui", "Quan Hung Tran", "Franck Dernoncourt", "Walter Chang" ]
Event coreference resolution is an important research problem with many applications. Despite the recent remarkable success of pre-trained language models, we argue that it is still highly beneficial to utilize symbolic features for the task. However, as the input for coreference resolution typically comes from upstrea...
2021.naacl-main.274
10.18653/v1/2021.naacl-main.274
null
2104.01697
title_snapshot
2021.naacl-main.275
Multi-Style Transfer with Discriminative Feedback on Disjoint Corpus
https://aclanthology.org/2021.naacl-main.275/
[ "Navita Goyal", "Balaji Vasan Srinivasan", "Anandhavelu N", "Abhilasha Sancheti" ]
Style transfer has been widely explored in natural language generation with non-parallel corpus by directly or indirectly extracting a notion of style from source and target domain corpus. A common shortcoming of existing approaches is the prerequisite of joint annotations across all the stylistic dimensions under cons...
2021.naacl-main.275
10.18653/v1/2021.naacl-main.275
null
2010.11578
title_snapshot
2021.naacl-main.276
FUDGE: Controlled Text Generation With Future Discriminators
https://aclanthology.org/2021.naacl-main.276/
[ "Kevin Yang", "Dan Klein" ]
We propose Future Discriminators for Generation (FUDGE), a flexible and modular method for controlled text generation. Given a pre-existing model G for generating text from a distribution of interest, FUDGE enables conditioning on a desired attribute a (for example, formality) while requiring access only to G’s output ...
2021.naacl-main.276
10.18653/v1/2021.naacl-main.276
null
2104.05218
title_snapshot
2021.naacl-main.277
Controllable Text Simplification with Explicit Paraphrasing
https://aclanthology.org/2021.naacl-main.277/
[ "Mounica Maddela", "Fernando Alva-Manchego", "Wei Xu" ]
Text Simplification improves the readability of sentences through several rewriting transformations, such as lexical paraphrasing, deletion, and splitting. Current simplification systems are predominantly sequence-to-sequence models that are trained end-to-end to perform all these operations simultaneously. However, su...
2021.naacl-main.277
10.18653/v1/2021.naacl-main.277
null
2010.11004
title_snapshot
2021.naacl-main.278
Knowledge Graph Based Synthetic Corpus Generation for Knowledge-Enhanced Language Model Pre-training
https://aclanthology.org/2021.naacl-main.278/
[ "Oshin Agarwal", "Heming Ge", "Siamak Shakeri", "Rami Al-Rfou" ]
Prior work on Data-To-Text Generation, the task of converting knowledge graph (KG) triples into natural text, focused on domain-specific benchmark datasets. In this paper, however, we verbalize the entire English Wikidata KG, and discuss the unique challenges associated with a broad, open-domain, large-scale verbalizat...
2021.naacl-main.278
10.18653/v1/2021.naacl-main.278
null
2010.12688
title_snapshot
2021.naacl-main.279
Choose Your Own Adventure: Paired Suggestions in Collaborative Writing for Evaluating Story Generation Models
https://aclanthology.org/2021.naacl-main.279/
[ "Elizabeth Clark", "Noah A. Smith" ]
Story generation is an open-ended and subjective task, which poses a challenge for evaluating story generation models. We present Choose Your Own Adventure, a collaborative writing setup for pairwise model evaluation. Two models generate suggestions to people as they write a short story; we ask writers to choose one of...
2021.naacl-main.279
10.18653/v1/2021.naacl-main.279
null
null
null
2021.naacl-main.280
InfoXLM: An Information-Theoretic Framework for Cross-Lingual Language Model Pre-Training
https://aclanthology.org/2021.naacl-main.280/
[ "Zewen Chi", "Li Dong", "Furu Wei", "Nan Yang", "Saksham Singhal", "Wenhui Wang", "Xia Song", "Xian-Ling Mao", "Heyan Huang", "Ming Zhou" ]
In this work, we present an information-theoretic framework that formulates cross-lingual language model pre-training as maximizing mutual information between multilingual-multi-granularity texts. The unified view helps us to better understand the existing methods for learning cross-lingual representations. More import...
2021.naacl-main.280
10.18653/v1/2021.naacl-main.280
null
2007.07834
title_snapshot
2021.naacl-main.281
Context-Interactive Pre-Training for Document Machine Translation
https://aclanthology.org/2021.naacl-main.281/
[ "Pengcheng Yang", "Pei Zhang", "Boxing Chen", "Jun Xie", "Weihua Luo" ]
Document machine translation aims to translate the source sentence into the target language in the presence of additional contextual information. However, it typically suffers from a lack of doc-level bilingual data. To remedy this, here we propose a simple yet effective context-interactive pre-training approach, which...
2021.naacl-main.281
10.18653/v1/2021.naacl-main.281
null
null
null
2021.naacl-main.282
Code-Mixing on Sesame Street: Dawn of the Adversarial Polyglots
https://aclanthology.org/2021.naacl-main.282/
[ "Samson Tan", "Shafiq Joty" ]
Multilingual models have demonstrated impressive cross-lingual transfer performance. However, test sets like XNLI are monolingual at the example level. In multilingual communities, it is common for polyglots to code-mix when conversing with each other. Inspired by this phenomenon, we present two strong black-box advers...
2021.naacl-main.282
10.18653/v1/2021.naacl-main.282
null
2103.09593
title_snapshot
2021.naacl-main.283
X-METRA-ADA: Cross-lingual Meta-Transfer learning Adaptation to Natural Language Understanding and Question Answering
https://aclanthology.org/2021.naacl-main.283/
[ "Meryem M’hamdi", "Doo Soon Kim", "Franck Dernoncourt", "Trung Bui", "Xiang Ren", "Jonathan May" ]
Multilingual models, such as M-BERT and XLM-R, have gained increasing popularity, due to their zero-shot cross-lingual transfer learning capabilities. However, their generalization ability is still inconsistent for typologically diverse languages and across different benchmarks. Recently, meta-learning has garnered att...
2021.naacl-main.283
10.18653/v1/2021.naacl-main.283
null
2104.09696
title_snapshot
2021.naacl-main.284
Explicit Alignment Objectives for Multilingual Bidirectional Encoders
https://aclanthology.org/2021.naacl-main.284/
[ "Junjie Hu", "Melvin Johnson", "Orhan Firat", "Aditya Siddhant", "Graham Neubig" ]
Pre-trained cross-lingual encoders such as mBERT (Devlin et al., 2019) and XLM-R (Conneau et al., 2020) have proven impressively effective at enabling transfer-learning of NLP systems from high-resource languages to low-resource languages. This success comes despite the fact that there is no explicit objective to align...
2021.naacl-main.284
10.18653/v1/2021.naacl-main.284
null
2010.07972
title_snapshot
2021.naacl-main.285
Cross-lingual Cross-modal Pretraining for Multimodal Retrieval
https://aclanthology.org/2021.naacl-main.285/
[ "Hongliang Fei", "Tan Yu", "Ping Li" ]
Recent pretrained vision-language models have achieved impressive performance on cross-modal retrieval tasks in English. Their success, however, heavily depends on the availability of many annotated image-caption datasets for pretraining, where the texts are not necessarily in English. Although we can utilize machine t...
2021.naacl-main.285
10.18653/v1/2021.naacl-main.285
null
null
null
2021.naacl-main.286
Wikipedia Entities as Rendezvous across Languages: Grounding Multilingual Language Models by Predicting Wikipedia Hyperlinks
https://aclanthology.org/2021.naacl-main.286/
[ "Iacer Calixto", "Alessandro Raganato", "Tommaso Pasini" ]
Masked language models have quickly become the de facto standard when processing text. Recently, several approaches have been proposed to further enrich word representations with external knowledge sources such as knowledge graphs. However, these models are devised and evaluated in a monolingual setting only. In this w...
2021.naacl-main.286
10.18653/v1/2021.naacl-main.286
null
null
null
2021.naacl-main.287
multiPRover: Generating Multiple Proofs for Improved Interpretability in Rule Reasoning
https://aclanthology.org/2021.naacl-main.287/
[ "Swarnadeep Saha", "Prateek Yadav", "Mohit Bansal" ]
We focus on a type of linguistic formal reasoning where the goal is to reason over explicit knowledge in the form of natural language facts and rules (Clark et al., 2020). A recent work, named PRover (Saha et al., 2020), performs such reasoning by answering a question and also generating a proof graph that explains the...
2021.naacl-main.287
10.18653/v1/2021.naacl-main.287
null
2106.01354
title_snapshot
2021.naacl-main.288
Adaptable and Interpretable Neural MemoryOver Symbolic Knowledge
https://aclanthology.org/2021.naacl-main.288/
[ "Pat Verga", "Haitian Sun", "Livio Baldini Soares", "William Cohen" ]
Past research has demonstrated that large neural language models (LMs) encode surprising amounts of factual information: however, augmenting or modifying this information requires modifying a corpus and retraining, which is computationally expensive. To address this problem, we develop a neural LM that includes an inte...
2021.naacl-main.288
10.18653/v1/2021.naacl-main.288
null
2007.00849
title_judge
2021.naacl-main.289
CLEVR_HYP: A Challenge Dataset and Baselines for Visual Question Answering with Hypothetical Actions over Images
https://aclanthology.org/2021.naacl-main.289/
[ "Shailaja Keyur Sampat", "Akshay Kumar", "Yezhou Yang", "Chitta Baral" ]
Most existing research on visual question answering (VQA) is limited to information explicitly present in an image or a video. In this paper, we take visual understanding to a higher level where systems are challenged to answer questions that involve mentally simulating the hypothetical consequences of performing speci...
2021.naacl-main.289
10.18653/v1/2021.naacl-main.289
null
2104.05981
title_snapshot
2021.naacl-main.290
Refining Targeted Syntactic Evaluation of Language Models
https://aclanthology.org/2021.naacl-main.290/
[ "Benjamin Newman", "Kai-Siang Ang", "Julia Gong", "John Hewitt" ]
Targeted syntactic evaluation of subject-verb number agreement in English (TSE) evaluates language models’ syntactic knowledge using hand-crafted minimal pairs of sentences that differ only in the main verb’s conjugation. The method evaluates whether language models rate each grammatical sentence as more likely than it...
2021.naacl-main.290
10.18653/v1/2021.naacl-main.290
null
2104.09635
title_snapshot
2021.naacl-main.291
Universal Adversarial Attacks with Natural Triggers for Text Classification
https://aclanthology.org/2021.naacl-main.291/
[ "Liwei Song", "Xinwei Yu", "Hsuan-Tung Peng", "Karthik Narasimhan" ]
Recent work has demonstrated the vulnerability of modern text classifiers to universal adversarial attacks, which are input-agnostic sequences of words added to text processed by classifiers. Despite being successful, the word sequences produced in such attacks are often ungrammatical and can be easily distinguished fr...
2021.naacl-main.291
10.18653/v1/2021.naacl-main.291
null
2005.00174
title_snapshot
2021.naacl-main.292
QuadrupletBERT: An Efficient Model For Embedding-Based Large-Scale Retrieval
https://aclanthology.org/2021.naacl-main.292/
[ "Peiyang Liu", "Sen Wang", "Xi Wang", "Wei Ye", "Shikun Zhang" ]
The embedding-based large-scale query-document retrieval problem is a hot topic in the information retrieval (IR) field. Considering that pre-trained language models like BERT have achieved great success in a wide variety of NLP tasks, we present a QuadrupletBERT model for effective and efficient retrieval in this pape...
2021.naacl-main.292
10.18653/v1/2021.naacl-main.292
null
null
null
2021.naacl-main.293
Dynamically Disentangling Social Bias from Task-Oriented Representations with Adversarial Attack
https://aclanthology.org/2021.naacl-main.293/
[ "Liwen Wang", "Yuanmeng Yan", "Keqing He", "Yanan Wu", "Weiran Xu" ]
Representation learning is widely used in NLP for a vast range of tasks. However, representations derived from text corpora often reflect social biases. This phenomenon is pervasive and consistent across different neural models, causing serious concern. Previous methods mostly rely on a pre-specified, user-provided dir...
2021.naacl-main.293
10.18653/v1/2021.naacl-main.293
null
null
null
2021.naacl-main.294
An Empirical Investigation of Bias in the Multimodal Analysis of Financial Earnings Calls
https://aclanthology.org/2021.naacl-main.294/
[ "Ramit Sawhney", "Arshiya Aggarwal", "Rajiv Ratn Shah" ]
Volatility prediction is complex due to the stock market’s stochastic nature. Existing research focuses on the textual elements of financial disclosures like earnings calls transcripts to forecast stock volatility and risk, but ignores the rich acoustic features in the company executives’ speech. Recently, new multimod...
2021.naacl-main.294
10.18653/v1/2021.naacl-main.294
null
null
null
2021.naacl-main.295
Beyond Fair Pay: Ethical Implications of NLP Crowdsourcing
https://aclanthology.org/2021.naacl-main.295/
[ "Boaz Shmueli", "Jan Fell", "Soumya Ray", "Lun-Wei Ku" ]
The use of crowdworkers in NLP research is growing rapidly, in tandem with the exponential increase in research production in machine learning and AI. Ethical discussion regarding the use of crowdworkers within the NLP research community is typically confined in scope to issues related to labor conditions such as fair ...
2021.naacl-main.295
10.18653/v1/2021.naacl-main.295
null
2104.10097
title_snapshot
2021.naacl-main.296
On Transferability of Bias Mitigation Effects in Language Model Fine-Tuning
https://aclanthology.org/2021.naacl-main.296/
[ "Xisen Jin", "Francesco Barbieri", "Brendan Kennedy", "Aida Mostafazadeh Davani", "Leonardo Neves", "Xiang Ren" ]
Fine-tuned language models have been shown to exhibit biases against protected groups in a host of modeling tasks such as text classification and coreference resolution. Previous works focus on detecting these biases, reducing bias in data representations, and using auxiliary training objectives to mitigate bias during...
2021.naacl-main.296
10.18653/v1/2021.naacl-main.296
null
2010.12864
title_snapshot
2021.naacl-main.297
Case Study: Deontological Ethics in NLP
https://aclanthology.org/2021.naacl-main.297/
[ "Shrimai Prabhumoye", "Brendon Boldt", "Ruslan Salakhutdinov", "Alan W Black" ]
Recent work in natural language processing (NLP) has focused on ethical challenges such as understanding and mitigating bias in data and algorithms; identifying objectionable content like hate speech, stereotypes and offensive language; and building frameworks for better system design and data handling practices. Howev...
2021.naacl-main.297
10.18653/v1/2021.naacl-main.297
null
2010.04658
title_snapshot
2021.naacl-main.298
Privacy Regularization: Joint Privacy-Utility Optimization in LanguageModels
https://aclanthology.org/2021.naacl-main.298/
[ "Fatemehsadat Mireshghallah", "Huseyin Inan", "Marcello Hasegawa", "Victor Rühle", "Taylor Berg-Kirkpatrick", "Robert Sim" ]
Neural language models are known to have a high capacity for memorization of training samples. This may have serious privacy im- plications when training models on user content such as email correspondence. Differential privacy (DP), a popular choice to train models with privacy guarantees, comes with significant costs...
2021.naacl-main.298
10.18653/v1/2021.naacl-main.298
null
2103.07567
title_judge
2021.naacl-main.299
On the Impact of Random Seeds on the Fairness of Clinical Classifiers
https://aclanthology.org/2021.naacl-main.299/
[ "Silvio Amir", "Jan-Willem van de Meent", "Byron Wallace" ]
Recent work has shown that fine-tuning large networks is surprisingly sensitive to changes in random seed(s). We explore the implications of this phenomenon for model fairness across demographic groups in clinical prediction tasks over electronic health records (EHR) in MIMIC-III —— the standard dataset in clinical NLP...
2021.naacl-main.299
10.18653/v1/2021.naacl-main.299
null
2104.06338
title_snapshot
2021.naacl-main.300
Topic Model or Topic Twaddle? Re-evaluating Semantic Interpretability Measures
https://aclanthology.org/2021.naacl-main.300/
[ "Caitlin Doogan", "Wray Buntine" ]
When developing topic models, a critical question that should be asked is: How well will this model work in an applied setting? Because standard performance evaluation of topic interpretability uses automated measures modeled on human evaluation tests that are dissimilar to applied usage, these models’ generalizability...
2021.naacl-main.300
10.18653/v1/2021.naacl-main.300
null
null
null