polygraf-ai/ACL-abstracts · Datasets at Hugging Face

inproceedings

kong-etal-2022-dropmix

{D}rop{M}ix: A Textual Data Augmentation Combining Dropout with Mixup

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.57/

Kong, Fanshuang and Zhang, Richong and Guo, Xiaohui and Mensah, Samuel and Mao, Yongyi

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

890--899

Overfitting is a notorious problem when there is insufficient data to train deep neural networks in machine learning tasks. Data augmentation regularization methods such as Dropout, Mixup, and their enhanced variants are effective and prevalent, and achieve promising performance to overcome overfitting. However, in text learning, most of the existing regularization approaches merely adopt ideas from computer vision without considering the importance of dimensionality in natural language processing. In this paper, we argue that the property is essential to overcome overfitting in text learning. Accordingly, we present a saliency map informed textual data augmentation and regularization framework, which combines Dropout and Mixup, namely DropMix, to mitigate the overfitting problem in text learning. In addition, we design a procedure that drops and patches fine grained shapes of the saliency map under the DropMix framework to enhance regularization. Empirical studies confirm the effectiveness of the proposed approach on 12 text classification tasks.

null

10.18653/v1/2022.emnlp-main.57

null

27,187

inproceedings

eirew-etal-2022-cross

Cross-document Event Coreference Search: Task, Dataset and Modeling

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.58/

Eirew, Alon and Caciularu, Avi and Dagan, Ido

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

900--913

The task of Cross-document Coreference Resolution has been traditionally formulated as requiring to identify all coreference links across a given set of documents. We propose an appealing, and often more applicable, complementary set up for the task {--} Cross-document Coreference Search, focusing in this paper on event coreference. Concretely, given a mention in context of an event of interest, considered as a query, the task is to find all coreferring mentions for the query event in a large document collection. To support research on this task, we create a corresponding dataset, which is derived from Wikipedia while leveraging annotations in the available Wikipedia Event Coreferecene dataset (WEC-Eng). Observing that the coreference search setup is largely analogous to the setting of Open Domain Question Answering, we adapt the prominent Deep Passage Retrieval (DPR) model to our setting, as an appealing baseline. Finally, we present a novel model that integrates a powerful coreference scoring scheme into the DPR architecture, yielding improved performance.

null

10.18653/v1/2022.emnlp-main.58

null

27,188

inproceedings

li-etal-2022-virt

{VIRT}: Improving Representation-based Text Matching via Virtual Interaction

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.59/

Li, Dan and Yang, Yang and Tang, Hongyin and Liu, Jiahao and Wang, Qifan and Wang, Jingang and Xu, Tong and Wu, Wei and Chen, Enhong

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

914--925

Text matching is a fundamental research problem in natural language understanding. Interaction-based approaches treat the text pair as a single sequence and encode it through cross encoders, while representation-based models encode the text pair independently with siamese or dual encoders. Interaction-based models require dense computations and thus are impractical in real-world applications. Representation-based models have become the mainstream paradigm for efficient text matching. However, these models suffer from severe performance degradation due to the lack of interactions between the pair of texts. To remedy this, we propose a Virtual InteRacTion mechanism (VIRT) for improving representation-based text matching while maintaining its efficiency. In particular, we introduce an interactive knowledge distillation module that is only applied during training. It enables deep interaction between texts by effectively transferring knowledge from the interaction-based model. A light interaction strategy is designed to fully leverage the learned interactive knowledge. Experimental results on six text matching benchmarks demonstrate the superior performance of our method over several state-of-the-art representation-based models. We further show that VIRT can be integrated into existing methods as plugins to lift their performances.

null

10.18653/v1/2022.emnlp-main.59

null

27,189

inproceedings

wang-etal-2022-maven

{MAVEN}-{ERE}: A Unified Large-scale Dataset for Event Coreference, Temporal, Causal, and Subevent Relation Extraction

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.60/

Wang, Xiaozhi and Chen, Yulin and Ding, Ning and Peng, Hao and Wang, Zimu and Lin, Yankai and Han, Xu and Hou, Lei and Li, Juanzi and Liu, Zhiyuan and Li, Peng and Zhou, Jie

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

926--941

The diverse relationships among real-world events, including coreference, temporal, causal, and subevent relations, are fundamental to understanding natural languages. However, two drawbacks of existing datasets limit event relation extraction (ERE) tasks: (1) Small scale. Due to the annotation complexity, the data scale of existing datasets is limited, which cannot well train and evaluate data-hungry models. (2) Absence of unified annotation. Different types of event relations naturally interact with each other, but existing datasets only cover limited relation types at once, which prevents models from taking full advantage of relation interactions. To address these issues, we construct a unified large-scale human-annotated ERE dataset MAVEN-ERE with improved annotation schemes. It contains 103,193 event coreference chains, 1,216,217 temporal relations, 57,992 causal relations, and 15,841 subevent relations, which is larger than existing datasets of all the ERE tasks by at least an order of magnitude. Experiments show that ERE on MAVEN-ERE is quite challenging, and considering relation interactions with joint learning can improve performances. The dataset and source codes can be obtained from https://github.com/THU-KEG/MAVEN-ERE.

null

10.18653/v1/2022.emnlp-main.60

null

27,190

inproceedings

mahapatra-etal-2022-entity

Entity Extraction in Low Resource Domains with Selective Pre-training of Large Language Models

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.61/

Mahapatra, Aniruddha and Nangi, Sharmila Reddy and Garimella, Aparna and N, Anandhavelu

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

942--951

Transformer-based language models trained on large natural language corpora have been very useful in downstream entity extraction tasks. However, they often result in poor performances when applied to domains that are different from those they are pretrained on. Continued pretraining using unlabeled data from target domains can help improve the performances of these language models on the downstream tasks. However, using all of the available unlabeled data for pretraining can be time-intensive; also, it can be detrimental to the performance of the downstream tasks, if the unlabeled data is not aligned with the data distribution for the target tasks. Previous works employed external supervision in the form of ontologies for selecting appropriate data samples for pretraining, but external supervision can be quite hard to obtain in low-resource domains. In this paper, we introduce effective ways to select data from unlabeled corpora of target domains for language model pretraining to improve the performances in target entity extraction tasks. Our data selection strategies do not require any external supervision. We conduct extensive experiments for the task of named entity recognition (NER) on seven different domains and show that language models pretrained on target domain unlabeled data obtained using our data selection strategies achieve better performances compared to those using data selection strategies in previous works that use external supervision. We also show that these pretrained language models using our data selection strategies outperform those pretrained on all of the available unlabeled target domain data.

null

10.18653/v1/2022.emnlp-main.61

null

27,191

inproceedings

wahle-etal-2022-large

How Large Language Models are Transforming Machine-Paraphrase Plagiarism

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.62/

Wahle, Jan Philip and Ruas, Terry and Kirstein, Frederic and Gipp, Bela

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

952--963

The recent success of large language models for text generation poses a severe threat to academic integrity, as plagiarists can generate realistic paraphrases indistinguishable from original work.However, the role of large autoregressive models in generating machine-paraphrased plagiarism and their detection is still incipient in the literature.This work explores T5 and GPT3 for machine-paraphrase generation on scientific articles from arXiv, student theses, and Wikipedia.We evaluate the detection performance of six automated solutions and one commercial plagiarism detection software and perform a human study with 105 participants regarding their detection performance and the quality of generated examples.Our results suggest that large language models can rewrite text humans have difficulty identifying as machine-paraphrased (53{\%} mean acc.).Human experts rate the quality of paraphrases generated by GPT-3 as high as original texts (clarity 4.0/5, fluency 4.2/5, coherence 3.8/5).The best-performing detection model (GPT-3) achieves 66{\%} F1-score in detecting paraphrases.We make our code, data, and findings publicly available to facilitate the development of detection solutions.

null

10.18653/v1/2022.emnlp-main.62

null

27,192

inproceedings

reid-etal-2022-m2d2

{M}2{D}2: A Massively Multi-Domain Language Modeling Dataset

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.63/

Reid, Machel and Zhong, Victor and Gururangan, Suchin and Zettlemoyer, Luke

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

964--975

We present M2D2, a fine-grained, massively multi-domain corpus for studying domain adaptation in language models (LMs). M2D2 consists of 8.5B tokens and spans 145 domains extracted from Wikipedia and Semantic Scholar. Using ontologies derived from Wikipedia and ArXiv categories, we organize the domains in each data source into 22 groups. This two-level hierarchy enables the study of relationships between domains and their effects on in- and out-of-domain performance after adaptation. We also present a number of insights into the nature of effective domain adaptation in LMs, as examples of the new types of studies M2D2 enables. To improve in-domain performance, we show the benefits of adapting the LM along a domain hierarchy; adapting to smaller amounts of fine-grained domain-specific data can lead to larger in-domain performance gains than larger amounts of weakly relevant data. We further demonstrate a trade-off between in-domain specialization and out-of-domain generalization within and across ontologies, as well as a strong correlation between out-of-domain performance and lexical overlap between domains.

null

10.18653/v1/2022.emnlp-main.63

null

27,193

inproceedings

bastings-etal-2022-will

{\textquotedblleft}Will You Find These Shortcuts?{\textquotedblright} A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.64/

Bastings, Jasmijn and Ebert, Sebastian and Zablotskaia, Polina and Sandholm, Anders and Filippova, Katja

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

976--991

Feature attribution a.k.a. input salience methods which assign an importance score to a feature are abundant but may produce surprisingly different results for the same model on the same input. While differences are expected if disparate definitions of importance are assumed, most methods claim to provide faithful attributions and point at the features most relevant for a model`s prediction. Existing work on faithfulness evaluation is not conclusive and does not provide a clear answer as to how different methods are to be compared.Focusing on text classification and the model debugging scenario, our main contribution is a protocol for faithfulness evaluation that makes use of partially synthetic data to obtain ground truth for feature importance ranking. Following the protocol, we do an in-depth analysis of four standard salience method classes on a range of datasets and lexical shortcuts for BERT and LSTM models. We demonstrate that some of the most popular method configurations provide poor results even for simple shortcuts while a method judged to be too simplistic works remarkably well for BERT.

null

10.18653/v1/2022.emnlp-main.64

null

27,194

inproceedings

zhang-feng-2022-information

Information-Transport-based Policy for Simultaneous Translation

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.65/

Zhang, Shaolei and Feng, Yang

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

992--1013

Simultaneous translation (ST) outputs translation while receiving the source inputs, and hence requires a policy to determine whether to translate a target token or wait for the next source token. The major challenge of ST is that each target token can only be translated based on the current received source tokens, where the received source information will directly affect the translation quality. So naturally, how much source information is received for the translation of the current target token is supposed to be the pivotal evidence for the ST policy to decide between translating and waiting. In this paper, we treat the translation as information transport from source to target and accordingly propose an Information-Transport-based Simultaneous Translation (ITST). ITST quantifies the transported information weight from each source token to the current target token, and then decides whether to translate the target token according to its accumulated received information. Experiments on both text-to-text ST and speech-to-text ST (a.k.a., streaming speech translation) tasks show that ITST outperforms strong baselines and achieves state-of-the-art performance.

null

10.18653/v1/2022.emnlp-main.65

null

27,195

inproceedings

li-etal-2022-learning-adapt

Learning to Adapt to Low-Resource Paraphrase Generation

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.66/

Li, Zhigen and Wang, Yanmeng and Fan, Rizhao and Wang, Ye and Li, Jianfeng and Wang, Shaojun

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1014--1022

Paraphrase generation is a longstanding NLP task and achieves great success with the aid of large corpora. However, transferring a paraphrasing model to another domain encounters the problem of domain shifting especially when the data is sparse. At the same time, widely using large pre-trained language models (PLMs) faces the overfitting problem when training on scarce labeled data. To mitigate these two issues, we propose, LAPA, an effective adapter for PLMs optimized by meta-learning. LAPA has three-stage training on three types of related resources to solve this problem: 1. pre-training PLMs on unsupervised corpora, 2. inserting an adapter layer and meta-training on source domain labeled data, and 3. fine-tuning adapters on a small amount of target domain labeled data. This method enables paraphrase generation models to learn basic language knowledge first, then learn the paraphrasing task itself later, and finally adapt to the target task. Our experimental results demonstrate that LAPA achieves state-of-the-art in supervised, unsupervised, and low-resource settings on three benchmark datasets. With only 2{\%} of trainable parameters and 1{\%} labeled data of the target task, our approach can achieve a competitive performance with previous work.

null

10.18653/v1/2022.emnlp-main.66

null

27,196

inproceedings

gu-etal-2022-distributional

A Distributional Lens for Multi-Aspect Controllable Text Generation

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.67/

Gu, Yuxuan and Feng, Xiaocheng and Ma, Sicheng and Zhang, Lingyuan and Gong, Heng and Qin, Bing

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1023--1043

Multi-aspect controllable text generation is a more challenging and practical task than single-aspect control. Existing methods achieve complex multi-aspect control by fusing multiple controllers learned from single-aspect, but suffer from attribute degeneration caused by the mutual interference of these controllers. To address this, we provide observations on attribute fusion from a distributional perspective and propose to directly search for the intersection areas of multiple attribute distributions as their combination for generation. Our method first estimates the attribute space with an autoencoder structure. Afterward, we iteratively approach the intersections by jointly minimizing distances to points representing different attributes. Finally, we map them to attribute-relevant sentences with a prefix-tuning-based decoder. Experiments on the three-aspect control task, including sentiment, topic, and detoxification aspects, reveal that our method outperforms several strong baselines on attribute relevance and text quality and achieves the SOTA. Further analysis also supplies some explanatory support for the effectiveness of our approach.

null

10.18653/v1/2022.emnlp-main.67

null

27,197

inproceedings

li-etal-2022-elmer

{ELMER}: A Non-Autoregressive Pre-trained Language Model for Efficient and Effective Text Generation

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.68/

Li, Junyi and Tang, Tianyi and Zhao, Wayne Xin and Nie, Jian-Yun and Wen, Ji-Rong

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1044--1058

We study the text generation task under the approach of pre-trained language models (PLMs). Typically, an auto-regressive (AR) method is adopted for generating texts in a token-by-token manner. Despite many advantages of AR generation, it usually suffers from inefficient inference. Therefore, non-autoregressive (NAR) models are proposed to generate all target tokens simultaneously. However, NAR models usually generate texts of lower quality due to the absence of token dependency in the output text. In this paper, we propose ELMER: an efficient and effective PLM for NAR text generation to explicitly model the token dependency during NAR generation. By leveraging the early exit technique, ELMER enables the token generations at different layers, according to their prediction confidence (a more confident token will exit at a lower layer). Besides, we propose a novel pre-training objective, Layer Permutation Language Modeling, to pre-train ELMER by permuting the exit layer for each token in sequences. Experiments on three text generation tasks show that ELMER significantly outperforms NAR models and further narrows the performance gap with AR PLMs (ELMER (29.92) vs BART (30.61) ROUGE-L in XSUM) while achieving over 10 times inference speedup.

null

10.18653/v1/2022.emnlp-main.68

null

27,198

inproceedings

chen-etal-2022-multilingual

Multilingual Relation Classification via Efficient and Effective Prompting

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.69/

Chen, Yuxuan and Harbecke, David and Hennig, Leonhard

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1059--1075

Prompting pre-trained language models has achieved impressive performance on various NLP tasks, especially in low data regimes. Despite the success of prompting in monolingual settings, applying prompt-based methods in multilingual scenarios has been limited to a narrow set of tasks, due to the high cost of handcrafting multilingual prompts. In this paper, we present the first work on prompt-based multilingual relation classification (RC), by introducing an efficient and effective method that constructs prompts from relation triples and involves only minimal translation for the class labels. We evaluate its performance in fully supervised, few-shot and zero-shot scenarios, and analyze its effectiveness across 14 languages, prompt variants, and English-task training in cross-lingual settings. We find that in both fully supervised and few-shot scenarios, our prompt method beats competitive baselines: fine-tuning XLM-R{\_}EM and null prompts. It also outperforms the random baseline by a large margin in zero-shot experiments. Our method requires little in-language knowledge and can be used as a strong baseline for similar multilingual classification tasks.

null

10.18653/v1/2022.emnlp-main.69

null

27,199

inproceedings

sawatphol-etal-2022-topic

Topic-Regularized Authorship Representation Learning

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.70/

Sawatphol, Jitkapat and Chaiwong, Nonthakit and Udomcharoenchaikit, Can and Nutanong, Sarana

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1076--1082

Authorship attribution is a task that aims to identify the author of a given piece of writing. We aim to develop a generalized solution that can handle a large number of texts from authors and topics unavailable in training data. Previous studies have proposed strategies to address only either unseen authors or unseen topics. Authorship representation learning has been shown to work in open-set environments with a large number of unseen authors but has not been explicitly designed for cross-topic environments at the same time. To handle a large number of unseen authors and topics, we propose Authorship Representation Regularization (ARR), a distillation framework that creates authorship representation with reduced reliance on topic-specific information. To assess the performance of our framework, we also propose a cross-topic-open-set evaluation method. Our proposed method has improved performances in the cross-topic-open set setup over baselines in 4 out of 6 cases.

null

10.18653/v1/2022.emnlp-main.70

null

27,200

inproceedings

hogan-etal-2022-fine

Fine-grained Contrastive Learning for Relation Extraction

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.71/

Hogan, William and Li, Jiacheng and Shang, Jingbo

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1083--1095

Recent relation extraction (RE) works have shown encouraging improvements by conducting contrastive learning on silver labels generated by distant supervision before fine-tuning on gold labels. Existing methods typically assume all these silver labels are accurate and treat them equally; however, distant supervision is inevitably noisy{--}some silver labels are more reliable than others. In this paper, we propose fine-grained contrastive learning (FineCL) for RE, which leverages fine-grained information about which silver labels are and are not noisy to improve the quality of learned relationship representations for RE. We first assess the quality of silver labels via a simple and automatic approach we call {\textquotedblleft}learning order denoising,{\textquotedblright} where we train a language model to learn these relations and record the order of learned training instances. We show that learning order largely corresponds to label accuracy{--}early-learned silver labels have, on average, more accurate labels than later-learned silver labels. Then, during pre-training, we increase the weights of accurate labels within a novel contrastive learning objective. Experiments on several RE benchmarks show that FineCL makes consistent and significant performance gains over state-of-the-art methods.

null

10.18653/v1/2022.emnlp-main.71

null

27,201

inproceedings

li-etal-2022-curriculum

Curriculum Prompt Learning with Self-Training for Abstractive Dialogue Summarization

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.72/

Li, Changqun and Wang, Linlin and Lin, Xin and de Melo, Gerard and He, Liang

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1096--1106

Succinctly summarizing dialogue is a task of growing interest, but inherent challenges, such as insufficient training data and low information density impede our ability to train abstractive models. In this work, we propose a novel curriculum-based prompt learning method with self-training to address these problems. Specifically, prompts are learned using a curriculum learning strategy that gradually increases the degree of prompt perturbation, thereby improving the dialogue understanding and modeling capabilities of our model. Unlabeled dialogue is incorporated by means of self-training so as to reduce the dependency on labeled data. We further investigate topic-aware prompts to better plan for the generation of summaries. Experiments confirm that our model substantially outperforms strong baselines and achieves new state-of-the-art results on the AMI and ICSI datasets. Human evaluations also show the superiority of our model with regard to the summary generation quality.

null

10.18653/v1/2022.emnlp-main.72

null

27,202

inproceedings

gera-etal-2022-zero

Zero-Shot Text Classification with Self-Training

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.73/

Gera, Ariel and Halfon, Alon and Shnarch, Eyal and Perlitz, Yotam and Ein-Dor, Liat and Slonim, Noam

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1107--1119

Recent advances in large pretrained language models have increased attention to zero-shot text classification. In particular, models finetuned on natural language inference datasets have been widely adopted as zero-shot classifiers due to their promising results and off-the-shelf availability. However, the fact that such models are unfamiliar with the target task can lead to instability and performance issues. We propose a plug-and-play method to bridge this gap using a simple self-training approach, requiring only the class names along with an unlabeled dataset, and without the need for domain expertise or trial and error. We show that fine-tuning the zero-shot classifier on its most confident predictions leads to significant performance gains across a wide range of text classification tasks, presumably since self-training adapts the zero-shot model to the task at hand.

null

10.18653/v1/2022.emnlp-main.73

null

27,203

inproceedings

santosh-etal-2022-deconfounding

Deconfounding Legal Judgment Prediction for {E}uropean Court of Human Rights Cases Towards Better Alignment with Experts

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.74/

T.y.s.s, Santosh and Xu, Shanshan and Ichim, Oana and Grabmair, Matthias

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1120--1138

This work demonstrates that Legal Judgement Prediction systems without expert-informed adjustments can be vulnerable to shallow, distracting surface signals that arise from corpus construction, case distribution, and confounding factors. To mitigate this, we use domain expertise to strategically identify statistically predictive but legally irrelevant information. We adopt adversarial training to prevent the system from relying on it. We evaluate our deconfounded models by employing interpretability techniques and comparing to expert annotations. Quantitative experiments and qualitative analysis show that our deconfounded model consistently aligns better with expert rationales than baselines trained for prediction only. We further contribute a set of reference expert annotations to the validation and testing partitions of an existing benchmark dataset of European Court of Human Rights cases.

null

10.18653/v1/2022.emnlp-main.74

null

27,204

inproceedings

wang-etal-2022-squality

{SQ}u{ALITY}: Building a Long-Document Summarization Dataset the Hard Way

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.75/

Wang, Alex and Pang, Richard Yuanzhe and Chen, Angelica and Phang, Jason and Bowman, Samuel R.

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1139--1156

Summarization datasets are often assembled either by scraping naturally occurring public-domain summaries{---}which are nearly always in difficult-to-work-with technical domains{---}or by using approximate heuristics to extract them from everyday text{---}which frequently yields unfaithful summaries. In this work, we turn to a slower but more straightforward approach to developing summarization benchmark data: We hire highly-qualified contractors to read stories and write original summaries from scratch. To amortize reading time, we collect five summaries per document, with the first giving an overview and the subsequent four addressing specific questions. We use this protocol to collect SQuALITY, a dataset of question-focused summaries built on the same public-domain short stories as the multiple-choice dataset QuALITY (Pang et al., 2021). Experiments with state-of-the-art summarization systems show that our dataset is challenging and that existing automatic evaluation metrics are weak indicators of quality.

null

10.18653/v1/2022.emnlp-main.75

null

27,205

inproceedings

ye-etal-2022-metaassist

{M}eta{ASSIST}: Robust Dialogue State Tracking with Meta Learning

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.76/

Ye, Fanghua and Wang, Xi and Huang, Jie and Li, Shenghui and Stern, Samuel and Yilmaz, Emine

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1157--1169

Existing dialogue datasets contain lots of noise in their state annotations. Such noise can hurt model training and ultimately lead to poor generalization performance. A general framework named ASSIST has recently been proposed to train robust dialogue state tracking (DST) models. It introduces an auxiliary model to generate pseudo labels for the noisy training set. These pseudo labels are combined with vanilla labels by a common fixed weighting parameter to train the primary DST model. Notwithstanding the improvements of ASSIST on DST, tuning the weighting parameter is challenging. Moreover, a single parameter shared by all slots and all instances may be suboptimal. To overcome these limitations, we propose a meta learning-based framework MetaASSIST to adaptively learn the weighting parameter. Specifically, we propose three schemes with varying degrees of flexibility, ranging from slot-wise to both slot-wise and instance-wise, to convert the weighting parameter into learnable functions. These functions are trained in a meta-learning manner by taking the validation set as meta data. Experimental results demonstrate that all three schemes can achieve competitive performance. Most impressively, we achieve a state-of-the-art joint goal accuracy of 80.10{\%} on MultiWOZ 2.4.

null

10.18653/v1/2022.emnlp-main.76

null

27,206

inproceedings

baziotis-etal-2022-multilingual

Multilingual Machine Translation with Hyper-Adapters

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.77/

Baziotis, Christos and Artetxe, Mikel and Cross, James and Bhosale, Shruti

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1170--1185

Multilingual machine translation suffers from negative interference across languages. A common solution is to relax parameter sharing with language-specific modules like adapters. However, adapters of related languages are unable to transfer information, and their total number of parameters becomes prohibitively expensive as the number of languages grows. In this work, we overcome these drawbacks using hyper-adapters {--} hyper-networks that generate adapters from language and layer embeddings. While past work had poor results when scaling hyper-networks, we propose a rescaling fix that significantly improves convergence and enables training larger hyper-networks. We find that hyper-adapters are more parameter efficient than regular adapters, reaching the same performance with up to 12 times less parameters. When using the same number of parameters and FLOPS, our approach consistently outperforms regular adapters. Also, hyper-adapters converge faster than alternative approaches and scale better than regular dense networks. Our analysis shows that hyper-adapters learn to encode language relatedness, enabling positive transfer across languages.

null

10.18653/v1/2022.emnlp-main.77

null

27,207

inproceedings

yang-etal-2022-z

{Z}-{L}a{VI}: Zero-Shot Language Solver Fueled by Visual Imagination

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.78/

Yang, Yue and Yao, Wenlin and Zhang, Hongming and Wang, Xiaoyang and Yu, Dong and Chen, Jianshu

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1186--1203

Large-scale pretrained language models have made significant advances in solving downstream language understanding tasks. However, they generally suffer from reporting bias, the phenomenon describing the lack of explicit commonsense knowledge in written text, e.g., {\textquotedblright}an orange is orange{\textquotedblright}. To overcome this limitation, we develop a novel approach, Z-LaVI, to endow language models with visual imagination capabilities. Specifically, we leverage two complementary types of {\textquotedblright}imaginations{\textquotedblright}: (i) recalling existing images through retrieval and (ii) synthesizing nonexistent images via text-to-image generation. Jointly exploiting the language inputs and the imagination, a pretrained vision-language model (e.g., CLIP) eventually composes a zero-shot solution to the original language tasks. Notably, fueling language models with imagination can effectively leverage visual knowledge to solve plain language tasks. In consequence, Z-LaVI consistently improves the zero-shot performance of existing language models across a diverse set of language tasks.

null

10.18653/v1/2022.emnlp-main.78

null

27,208

inproceedings

lal-etal-2022-using

Using Commonsense Knowledge to Answer Why-Questions

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.79/

Lal, Yash Kumar and Tandon, Niket and Aggarwal, Tanvi and Liu, Horace and Chambers, Nathanael and Mooney, Raymond and Balasubramanian, Niranjan

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1204--1219

Answering questions in narratives about why events happened often requires commonsense knowledge external to the text. What aspects of this knowledge are available in large language models? What aspects can be made accessible via external commonsense resources? We study these questions in the context of answering questions in the TellMeWhy dataset using COMET as a source of relevant commonsense relations. We analyze the effects of model size (T5 and GPT3) along with methods of injecting knowledge (COMET) into these models. Results show that the largest models, as expected, yield substantial improvements over base models. Injecting external knowledge helps models of various sizes, but the amount of improvement decreases with larger model size. We also find that the format in which knowledge is provided is critical, and that smaller models benefit more from larger amounts of knowledge. Finally, we develop an ontology of knowledge types and analyze the relative coverage of the models across these categories.

null

10.18653/v1/2022.emnlp-main.79

null

27,209

inproceedings

ch-wang-etal-2022-affective

Affective Idiosyncratic Responses to Music

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.80/

CH-Wang, Sky and Li, Evan and Li, Oliver and Muresan, Smaranda and Yu, Zhou

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1220--1250

Affective responses to music are highly personal. Despite consensus that idiosyncratic factors play a key role in regulating how listeners emotionally respond to music, precisely measuring the marginal effects of these variables has proved challenging. To address this gap, we develop computational methods to measure affective responses to music from over 403M listener comments on a Chinese social music platform. Building on studies from music psychology in systematic and quasi-causal analyses, we test for musical, lyrical, contextual, demographic, and mental health effects that drive listener affective responses. Finally, motivated by the social phenomenon known as 网抑云 (wǎng-y{\`i}-y{\'u}n), we identify influencing factors of platform user self-disclosures, the social support they receive, and notable differences in discloser user activity.

null

10.18653/v1/2022.emnlp-main.80

null

27,210

inproceedings

dua-etal-2022-successive

Successive Prompting for Decomposing Complex Questions

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.81/

Dua, Dheeru and Gupta, Shivanshu and Singh, Sameer and Gardner, Matt

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1251--1265

Answering complex questions that require making latent decisions is a challenging task, especially when limited supervision is available. Recent works leverage the capabilities of large language models (LMs) to perform complex question answering in a few-shot setting by demonstrating how to output intermediate rationalizations while solving the complex question in a single pass. We introduce {\textquotedblleft}Successive Prompting{\textquotedblright} where, we iteratively break down a complex task into a simple task, solve it, and then repeat the process until we get the final solution. Successive prompting decouples the supervision for decomposing complex questions from the supervision for answering simple questions, allowing us to (1) have multiple opportunities to query in-context examples at each reasoning step (2) learn question decomposition separately from question answering, including using synthetic data, and (3) use bespoke (fine-tuned) components for reasoning steps where a large LM does not perform well. The intermediate supervision is typically manually written, which can be expensive to collect. We introduce a way to generate synthetic dataset which can be used to bootstrap model`s ability to decompose and answer intermediate questions. Our best model (with successive prompting) achieves an improvement in F1 of {\textasciitilde}5{\%} when compared with a state-of-the-art model with synthetic augmentations and few-shot version of the DROP dataset.

null

10.18653/v1/2022.emnlp-main.81

null

27,211

inproceedings

jung-etal-2022-maieutic

Maieutic Prompting: Logically Consistent Reasoning with Recursive Explanations

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.82/

Jung, Jaehun and Qin, Lianhui and Welleck, Sean and Brahman, Faeze and Bhagavatula, Chandra and Le Bras, Ronan and Choi, Yejin

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1266--1279

Pre-trained language models (LMs) struggle with consistent reasoning; recently, prompting LMs to generate explanations that self-guide the inference has emerged as a promising direction to amend this. However, these approaches are fundamentally bounded by the correctness of explanations, which themselves are often noisy and inconsistent. In this work, we develop Maieutic Prompting, which aims to infer a correct answer to a question even from the unreliable generations of LM. Maieutic Prompting induces a tree of explanations abductively (e.g. X is true, because ...) and recursively, then frames the inference as a satisfiability problem over these explanations and their logical relations. We test Maieutic Prompting for true/false QA on three challenging benchmarks that require complex commonsense reasoning. Maieutic Prompting achieves up to 20{\%} better accuracy than state-of-the-art prompting methods, and as a fully unsupervised approach, performs competitively with supervised models. We also show that Maieutic Prompting improves robustness in inference while providing interpretable rationales.

null

10.18653/v1/2022.emnlp-main.82

null

27,212

inproceedings

zhang-etal-2022-danli

{DANLI}: Deliberative Agent for Following Natural Language Instructions

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.83/

Zhang, Yichi and Yang, Jianing and Pan, Jiayi and Storks, Shane and Devraj, Nikhil and Ma, Ziqiao and Yu, Keunwoo and Bao, Yuwei and Chai, Joyce

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1280--1298

Recent years have seen an increasing amount of work on embodied AI agents that can perform tasks by following human language instructions. However, most of these agents are reactive, meaning that they simply learn and imitate behaviors encountered in the training data. These reactive agents are insufficient for long-horizon complex tasks. To address this limitation, we propose a neuro-symbolic deliberative agent that, while following language instructions, proactively applies reasoning and planning based on its neural and symbolic representations acquired from past experience (e.g., natural language and egocentric vision). We show that our deliberative agent achieves greater than 70{\%} improvement over reactive baselines on the challenging TEACh benchmark. Moreover, the underlying reasoning and planning processes, together with our modular framework, offer impressive transparency and explainability to the behaviors of the agent. This enables an in-depth understanding of the agent`s capabilities, which shed light on challenges and opportunities for future embodied agents for instruction following. The code is available at https://github.com/sled-group/DANLI.

null

10.18653/v1/2022.emnlp-main.83

null

27,213

inproceedings

sun-xu-2022-tracing

Tracing Semantic Variation in Slang

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.84/

Sun, Zhewei and Xu, Yang

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1299--1313

The meaning of a slang term can vary in different communities. However, slang semantic variation is not well understood and under-explored in the natural language processing of slang. One existing view argues that slang semantic variation is driven by culture-dependent communicative needs. An alternative view focuses on slang`s social functions suggesting that the desire to foster semantic distinction may have led to the historical emergence of community-specific slang senses. We explore these theories using computational models and test them against historical slang dictionary entries, with a focus on characterizing regularity in the geographical variation of slang usages attested in the US and the UK over the past two centuries. We show that our models are able to predict the regional identity of emerging slang word meanings from historical slang records. We offer empirical evidence that both communicative need and semantic distinction play a role in the variation of slang meaning yet their relative importance fluctuates over the course of history. Our work offers an opportunity for incorporating historical cultural elements into the natural language processing of slang.

null

10.18653/v1/2022.emnlp-main.84

null

27,214

inproceedings

an-etal-2022-fine

Fine-grained Category Discovery under Coarse-grained supervision with Hierarchical Weighted Self-contrastive Learning

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.85/

An, Wenbin and Tian, Feng and Chen, Ping and Tang, Siliang and Zheng, Qinghua and Wang, QianYing

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1314--1323

Novel category discovery aims at adapting models trained on known categories to novel categories. Previous works only focus on the scenario where known and novel categories are of the same granularity.In this paper, we investigate a new practical scenario called Fine-grained Category Discovery under Coarse-grained supervision (FCDC). FCDC aims at discovering fine-grained categories with only coarse-grained labeled data, which can adapt models to categories of different granularity from known ones and reduce significant labeling cost. It is also a challenging task since supervised training on coarse-grained categories tends to focus on inter-class distance (distance between coarse-grained classes) but ignore intra-class distance (distance between fine-grained sub-classes) which is essential for separating fine-grained categories.Considering most current methods cannot transfer knowledge from coarse-grained level to fine-grained level, we propose a hierarchical weighted self-contrastive network by building a novel weighted self-contrastive module and combining it with supervised learning in a hierarchical manner.Extensive experiments on public datasets show both effectiveness and efficiency of our model over compared methods.

null

10.18653/v1/2022.emnlp-main.85

null

27,215

inproceedings

kim-etal-2022-plm

{PLM}-based World Models for Text-based Games

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.86/

Kim, Minsoo and Jung, Yeonjoon and Lee, Dohyeon and Hwang, Seung-won

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1324--1341

World models have improved the ability of reinforcement learning agents to operate in a sample efficient manner, by being trained to predict plausible changes in the underlying environment. As the core tasks of world models are future prediction and commonsense understanding, our claim is that pre-trained language models (PLMs) already provide a strong base upon which to build world models. Worldformer is a recently proposed world model for text-based game environments, based only partially on PLM and transformers. Our distinction is to fully leverage PLMs as actionable world models in text-based game environments, by reformulating generation as constrained decoding which decomposes actions into verb templates and objects. We show that our model improves future valid action prediction and graph change prediction. Additionally, we show that our model better reflects commonsense than standard PLM.

null

10.18653/v1/2022.emnlp-main.86

null

27,216

inproceedings

zhang-etal-2022-prompt-based

Prompt-Based Meta-Learning For Few-shot Text Classification

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.87/

Zhang, Haoxing and Zhang, Xiaofeng and Huang, Haibo and Yu, Lei

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1342--1357

Few-shot Text Classification predicts the semantic label of a given text with a handful of supporting instances. Current meta-learning methods have achieved satisfying results in various few-shot situations. Still, they often require a large amount of data to construct many few-shot tasks for meta-training, which is not practical in real-world few-shot scenarios. Prompt-tuning has recently proved to be another effective few-shot learner by bridging the gap between pre-train and downstream tasks. In this work, we closely combine the two promising few-shot learning methodologies in structure and propose a Prompt-Based Meta-Learning (PBML) model to overcome the above meta-learning problem by adding the prompting mechanism. PBML assigns label word learning to base-learners and template learning to meta-learner, respectively. Experimental results show state-of-the-art performance on four text classification datasets under few-shot settings, with higher accuracy and good robustness. We demonstrate through low-resource experiments that our method alleviates the shortcoming that meta-learning requires too much data for meta-training. In the end, we use the visualization to interpret and verify that the meta-learning framework can help the prompting method converge better. We release our code to reproduce our experiments.

null

10.18653/v1/2022.emnlp-main.87

null

27,217

inproceedings

bansal-etal-2022-well

How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.88/

Bansal, Hritik and Yin, Da and Monajatipoor, Masoud and Chang, Kai-Wei

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1358--1370

Text-to-image generative models have achieved unprecedented success in generating high-quality images based on natural language descriptions. However, it is shown that these models tend to favor specific social groups when prompted with neutral text descriptions (e.g., {\textquoteleft}a photo of a lawyer'). Following Zhao et al. (2021), we study the effect on the diversity of the generated images when adding \textit{ethical intervention} that supports equitable judgment (e.g., {\textquoteleft}if all individuals can be a lawyer irrespective of their gender') in the input prompts. To this end, we introduce an Ethical NaTural Language Interventions in Text-to-Image GENeration (ENTIGEN) benchmark dataset to evaluate the change in image generations conditional on ethical interventions across three social axes {--} gender, skin color, and culture. Through CLIP-based and human evaluation on minDALL.E, DALL.E-mini and Stable Diffusion, we find that the model generations cover diverse social groups while preserving the image quality. In some cases, the generations would be anti-stereotypical (e.g., models tend to create images with individuals that are perceived as man when fed with prompts about makeup) in the presence of ethical intervention. Preliminary studies indicate that a large change in the model predictions is triggered by certain phrases such as {\textquoteleft}irrespective of gender' in the context of gender bias in the ethical interventions. We release code and annotated data at https://github.com/Hritikbansal/entigen{\_}emnlp.

null

10.18653/v1/2022.emnlp-main.88

null

27,218

inproceedings

rungta-etal-2022-geographic

Geographic Citation Gaps in {NLP} Research

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.89/

Rungta, Mukund and Singh, Janvijay and Mohammad, Saif M. and Yang, Diyi

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1371--1383

In a fair world, people have equitable opportunities to education, to conduct scientific research, to publish, and to get credit for their work, regardless of where they live. However, it is common knowledge among researchers that a vast number of papers accepted at top NLP venues come from a handful of western countries and (lately) China; whereas, very few papers from Africa and South America get published. Similar disparities are also believed to exist for paper citation counts. In the spirit of {\textquotedblleft}what we do not measure, we cannot improve{\textquotedblright}, this work asks a series of questions on the relationship between geographical location and publication success (acceptance in top NLP venues and citation impact). We first created a dataset of 70,000 papers from the ACL Anthology, extracted their meta-information, andgenerated their citation network. We then show that not only are there substantial geographical disparities in paper acceptance and citation but also that these disparities persist even when controlling for a number of variables such as venue of publication and sub-field of NLP. Further, despite some steps taken by the NLP community to improve geographical diversity, we show that the disparity in publication metrics across locations is still on an increasing trend since the early 2000s. We release our code and dataset here: https://github.com/iamjanvijay/acl-cite-net

null

10.18653/v1/2022.emnlp-main.89

null

27,219

inproceedings

madaan-etal-2022-language

Language Models of Code are Few-Shot Commonsense Learners

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.90/

Madaan, Aman and Zhou, Shuyan and Alon, Uri and Yang, Yiming and Neubig, Graham

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1384--1403

We address the general task of structured commonsense reasoning: given a natural language input, the goal is to generate a graph such as an event or a reasoning-graph.To employ large language models (LMs) for this task, existing approaches {\textquoteleft}serialize' the output graph as a flat list of nodes and edges.Although feasible, these serialized graphs strongly deviate from the natural language corpora that LMs were pre-trained on, hindering LMs from generating them correctly. In this paper, we show that when we instead frame structured commonsense reasoning tasks as code generation tasks, pre-trained LMs of code are better structured commonsense reasoners than LMs of natural language, even when the downstream task does not involve source code at all.We demonstrate our approach across three diverse structured commonsense reasoning tasks. In all these natural language tasks, we show that using our approach, a code generation LM (codex) outperforms natural-LMs that are fine-tuned on the target task (T5) and other strong LMs such as GPT-3 in the few-shot setting.

null

10.18653/v1/2022.emnlp-main.90

null

27,220

inproceedings

hua-etal-2022-numerical

Numerical Optimizations for Weighted Low-rank Estimation on Language Models

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.91/

Hua, Ting and Hsu, Yen-Chang and Wang, Felicity and Lou, Qian and Shen, Yilin and Jin, Hongxia

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1404--1416

Singular value decomposition (SVD) is one of the most popular compression methods that approximate a target matrix with smaller matrices. However, standard SVD treats the parameters within the matrix with equal importance, which is a simple but unrealistic assumption. The parameters of a trained neural network model may affect the task performance unevenly, which suggests non-equal importance among the parameters. Compared to SVD, the decomposition method aware of parameter importance is the more practical choice in real cases. Unlike standard SVD, weighed value decomposition is a non-convex optimization problem that lacks a closed-form solution. We systematically investigated multiple optimization strategies to tackle the problem and examined our method by compressing Transformer-based language models.Further, we designed a metric to predict when the SVD may introduce a significant performance drop, for which our method can be a rescue strategy.The extensive evaluations demonstrate that our method can perform better than current SOTA methods in compressing Transformer-based language models.

null

10.18653/v1/2022.emnlp-main.91

null

27,221

inproceedings

lee-etal-2022-generative

Generative Multi-hop Retrieval

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.92/

Lee, Hyunji and Yang, Sohee and Oh, Hanseok and Seo, Minjoon

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1417--1436

A common practice for text retrieval is to use an encoder to map the documents and the query to a common vector space and perform a nearest neighbor search (NNS); multi-hop retrieval also often adopts the same paradigm, usually with a modification of iteratively reformulating the query vector so that it can retrieve different documents at each hop. However, such a bi-encoder approach has limitations in multi-hop settings; (1) the reformulated query gets longer as the number of hops increases, which further tightens the embedding bottleneck of the query vector, and (2) it is prone to error propagation. In this paper, we focus on alleviating these limitations in multi-hop settings by formulating the problem in a fully generative way. We propose an encoder-decoder model that performs multi-hop retrieval by simply generating the entire text sequences of the retrieval targets, which means the query and the documents interact in the language model`s parametric space rather than L2 or inner product space as in the bi-encoder approach. Our approach, Generative Multi-hop Retrieval (GMR), consistently achieves comparable or higher performance than bi-encoder models in five datasets while demonstrating superior GPU memory and storage footprint.

null

10.18653/v1/2022.emnlp-main.92

null

27,222

inproceedings

zhao-etal-2022-visual

Visual Spatial Description: Controlled Spatial-Oriented Image-to-Text Generation

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.93/

Zhao, Yu and Wei, Jianguo and Lin, ZhiChao and Sun, Yueheng and Zhang, Meishan and Zhang, Min

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1437--1449

Image-to-text tasks such as open-ended image captioning and controllable image description have received extensive attention for decades. Here we advance this line of work further, presenting Visual Spatial Description (VSD), a new perspective for image-to-text toward spatial semantics. Given an image and two objects inside it, VSD aims to produce one description focusing on the spatial perspective between the two objects. Accordingly, we annotate a dataset manually to facilitate the investigation of the newly-introduced task, and then build several benchmark encoder-decoder models by using VL-BART and VL-T5 as backbones. In addition, we investigate visual spatial relationship classification (VSRC) information into our model by pipeline and end-to-end architectures. Finally, we conduct experiments on our benchmark dataset to evaluate all our models. Results show that our models are awe-inspiring, offering accurate and human-like spatial-oriented text descriptions. Besides, VSRC has great potential for VSD, and the joint end-to-end architecture is the better choice for their integration. We will make the dataset and codes publicly available for research purposes.

null

10.18653/v1/2022.emnlp-main.93

null

27,223

inproceedings

wen-etal-2022-m3

{M}3: A Multi-View Fusion and Multi-Decoding Network for Multi-Document Reading Comprehension

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.94/

Wen, Liang and Wang, Houfeng and Luo, Yingwei and Wang, Xiaolin

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1450--1461

Multi-document reading comprehension task requires collecting evidences from different documents for answering questions. Previous research works either use the extractive modeling method to naively integrate the scores from different documents on the encoder side or use the generative modeling method to collect the clues from different documents on the decoder side individually. However, any single modeling method cannot make full of the advantages of both. In this work, we propose a novel method that tries to employ a multi-view fusion and multi-decoding mechanism to achieve it. For one thing, our approach leverages question-centered fusion mechanism and cross-attention mechanism to gather fine-grained fusion of evidence clues from different documents in the encoder and decoder concurrently. For another, our method simultaneously employs both the extractive decoding approach and the generative decoding method to effectively guide the training process. Compared with existing methods, our method can perform both extractive decoding and generative decoding independently and optionally. Our experiments on two mainstream multi-document reading comprehension datasets (Natural Questions and TriviaQA) demonstrate that our method can provide consistent improvements over previous state-of-the-art methods.

null

10.18653/v1/2022.emnlp-main.94

null

27,224

inproceedings

yu-etal-2022-coco

{COCO}-{DR}: Combating the Distribution Shift in Zero-Shot Dense Retrieval with Contrastive and Distributionally Robust Learning

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.95/

Yu, Yue and Xiong, Chenyan and Sun, Si and Zhang, Chao and Overwijk, Arnold

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1462--1479

We present a new zero-shot dense retrieval (ZeroDR) method, COCO-DR, to improve the generalization ability of dense retrieval by combating the distribution shifts between source training tasks and target scenarios. To mitigate the impact of document differences, COCO-DR continues pretraining the language model on the target corpora to adapt the model to target distributions via COtinuous COtrastive learning. To prepare for unseen target queries, COCO-DR leverages implicit Distributionally Robust Optimization (iDRO) to reweight samples from different source query clusters for improving model robustness over rare queries during fine-tuning. COCO-DR achieves superior average performance on BEIR, the zero-shot retrieval benchmark. At BERT{\_}Base scale, COCO-DR Base outperforms other ZeroDR models with 60x larger size. At BERT{\_}Large scale, COCO-DR Large outperforms the giant GPT-3 embedding model which has 500x more parameters. Our analysis shows the correlation between COCO-DR`s effectiveness in combating distribution shifts and improving zero-shot accuracy. Our code and model can be found at \url{https://github.com/OpenMatch/COCO-DR}.

null

10.18653/v1/2022.emnlp-main.95

null

27,225

inproceedings

ren-etal-2022-language

Language Model Pre-Training with Sparse Latent Typing

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.96/

Ren, Liliang and Zhang, Zixuan and Wang, Han and Voss, Clare and Zhai, ChengXiang and Ji, Heng

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1480--1494

Modern large-scale Pre-trained Language Models (PLMs) have achieved tremendous success on a wide range of downstream tasks. However, most of the LM pre-training objectives only focus on text reconstruction, but have not sought to learn latent-level interpretable representations of sentences. In this paper, we manage to push the language models to obtain a deeper understanding of sentences by proposing a new pre-training objective, Sparse Latent Typing, which enables the model to sparsely extract sentence-level keywords with diverse latent types. Experimental results show that our model is able to learn interpretable latent type categories in a self-supervised manner without using any external knowledge. Besides, the language model pre-trained with such an objective also significantly improves Information Extraction related downstream tasks in both supervised and few-shot settings. Our code is publicly available at https://github.com/renll/SparseLT.

null

10.18653/v1/2022.emnlp-main.96

null

27,226

inproceedings

durrani-etal-2022-transformation

On the Transformation of Latent Space in Fine-Tuned {NLP} Models

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.97/

Durrani, Nadir and Sajjad, Hassan and Dalvi, Fahim and Alam, Firoj

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1495--1516

We study the evolution of latent space in fine-tuned NLP models. Different from the commonly used probing-framework, we opt for an unsupervised method to analyze representations. More specifically, we discover latent concepts in the representational space using hierarchical clustering. We then use an alignment function to gauge the similarity between the latent space of a pre-trained model and its fine-tuned version. We use traditional linguistic concepts to facilitate our understanding and also study how the model space transforms towards task-specific information. We perform a thorough analysis, comparing pre-trained and fine-tuned models across three models and three downstream tasks. The notable findings of our work are: i) the latent space of the higher layers evolve towards task-specific concepts, ii) whereas the lower layers retain generic concepts acquired in the pre-trained model, iii) we discovered that some concepts in the higher layers acquire polarity towards the output class, and iv) that these concepts can be used for generating adversarial triggers.

null

10.18653/v1/2022.emnlp-main.97

null

27,227

inproceedings

mou-etal-2022-watch

Watch the Neighbors: A Unified K-Nearest Neighbor Contrastive Learning Framework for {OOD} Intent Discovery

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.98/

Mou, Yutao and He, Keqing and Wang, Pei and Wu, Yanan and Wang, Jingang and Wu, Wei and Xu, Weiran

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1517--1529

Discovering out-of-domain (OOD) intent is important for developing new skills in task-oriented dialogue systems. The key challenges lie in how to transfer prior in-domain (IND) knowledge to OOD clustering, as well as jointly learn OOD representations and cluster assignments. Previous methods suffer from in-domain overfitting problem, and there is a natural gap between representation learning and clustering objectives. In this paper, we propose a unified K-nearest neighbor contrastive learning framework to discover OOD intents. Specifically, for IND pre-training stage, we propose a KCL objective to learn inter-class discriminative features, while maintaining intra-class diversity, which alleviates the in-domain overfitting problem. For OOD clustering stage, we propose a KCC method to form compact clusters by mining true hard negative samples, which bridges the gap between clustering and representation learning. Extensive experiments on three benchmark datasets show that our method achieves substantial improvements over the state-of-the-art methods.

null

10.18653/v1/2022.emnlp-main.98

null

27,228

inproceedings

he-etal-2022-extracted

Extracted {BERT} Model Leaks More Information than You Think!

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.99/

He, Xuanli and Lyu, Lingjuan and Chen, Chen and Xu, Qiongkai

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1530--1537

The collection and availability of big data, combined with advances in pre-trained models (e.g. BERT), have revolutionized the predictive performance of natural language processing tasks. This allows corporations to provide machine learning as a service (MLaaS) by encapsulating fine-tuned BERT-based models as APIs. Due to significant commercial interest, there has been a surge of attempts to steal remote services via model extraction. Although previous works have made progress in defending against model extraction attacks, there has been little discussion on their performance in preventing privacy leakage. This work bridges this gap by launching an attribute inference attack against the extracted BERT model. Our extensive experiments reveal that model extraction can cause severe privacy leakage even when victim models are facilitated with state-of-the-art defensive strategies.

null

10.18653/v1/2022.emnlp-main.99

null

27,229

inproceedings

nikolaus-etal-2022-vision

Do Vision-and-Language Transformers Learn Grounded Predicate-Noun Dependencies?

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.100/

Nikolaus, Mitja and Salin, Emmanuelle and Ayache, Stephane and Fourtassi, Abdellah and Favre, Benoit

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1538--1555

Recent advances in vision-and-language modeling have seen the development of Transformer architectures that achieve remarkable performance on multimodal reasoning tasks.Yet, the exact capabilities of these black-box models are still poorly understood. While much of previous work has focused on studying their ability to learn meaning at the word-level, their ability to track syntactic dependencies between words has received less attention.We take a first step in closing this gap by creating a new multimodal task targeted at evaluating understanding of predicate-noun dependencies in a controlled setup.We evaluate a range of state-of-the-art models and find that their performance on the task varies considerably, with some models performing relatively well and others at chance level. In an effort to explain this variability, our analyses indicate that the quality (and not only sheer quantity) of pretraining data is essential. Additionally, the best performing models leverage fine-grained multimodal pretraining objectives in addition to the standard image-text matching objectives.This study highlights that targeted and controlled evaluations are a crucial step for a precise and rigorous test of the multimodal knowledge of vision-and-language models.

null

10.18653/v1/2022.emnlp-main.100

null

27,230

inproceedings

zaman-belinkov-2022-multilingual

A Multilingual Perspective Towards the Evaluation of Attribution Methods in Natural Language Inference

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.101/

Zaman, Kerem and Belinkov, Yonatan

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1556--1576

Most evaluations of attribution methods focus on the English language. In this work, we present a multilingual approach for evaluating attribution methods for the Natural Language Inference (NLI) task in terms of faithfulness and plausibility.First, we introduce a novel cross-lingual strategy to measure faithfulness based on word alignments, which eliminates the drawbacks of erasure-based evaluations.We then perform a comprehensive evaluation of attribution methods, considering different output mechanisms and aggregation methods.Finally, we augment the XNLI dataset with highlight-based explanations, providing a multilingual NLI dataset with highlights, to support future exNLP studies. Our results show that attribution methods performing best for plausibility and faithfulness are different.

null

10.18653/v1/2022.emnlp-main.101

null

27,231

inproceedings

imanigooghari-etal-2022-graph

Graph-Based Multilingual Label Propagation for Low-Resource Part-of-Speech Tagging

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.102/

Imani, Ayyoob and Severini, Silvia and Jalili Sabet, Masoud and Yvon, Fran{\c{cois and Sch{\"utze, Hinrich

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1577--1589

Part-of-Speech (POS) tagging is an important component of the NLP pipeline, but many low-resource languages lack labeled data for training. An established method for training a POS tagger in such a scenario is to create a labeled training set by transferring from high-resource languages. In this paper, we propose a novel method for transferring labels from multiple high-resource source to low-resource target languages. We formalize POS tag projection as graph-based label propagation. Given translations of a sentence in multiple languages, we create a graph with words as nodes and alignment links as edges by aligning words for all language pairs. We then propagate node labels from source to target using a Graph Neural Network augmented with transformer layers. We show that our propagation creates training sets that allow us to train POS taggers for a diverse set of languages. When combined with enhanced contextualized embeddings, our method achieves a new state-of-the-art for unsupervised POS tagging of low-resource languages.

null

10.18653/v1/2022.emnlp-main.102

null

27,232

inproceedings

wang-etal-2022-subeventwriter

{S}ubevent{W}riter: Iterative Sub-event Sequence Generation with Coherence Controller

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.103/

Wang, Zhaowei and Zhang, Hongming and Fang, Tianqing and Song, Yangqiu and Wong, Ginny and See, Simon

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1590--1604

In this paper, we propose a new task of sub-event generation for an unseen process to evaluate the understanding of the coherence of sub-event actions and objects. To solve the problem, we design SubeventWriter, a sub-event sequence generation framework with a coherence controller. Given an unseen process, the framework can iteratively construct the sub-event sequence by generating one sub-event at each iteration. We also design a very effective coherence controller to decode more coherent sub-events. As our extensive experiments and analysis indicate, SubeventWriter can generate more reliable and meaningful sub-event sequences for unseen processes.

null

10.18653/v1/2022.emnlp-main.103

null

27,233

inproceedings

inoue-etal-2022-infinite

Infinite {SCAN}: An Infinite Model of Diachronic Semantic Change

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.104/

Inoue, Seiichi and Komachi, Mamoru and Ogiso, Toshinobu and Takamura, Hiroya and Mochihashi, Daichi

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1605--1616

In this study, we propose a Bayesian model that can jointly estimate the number of senses of words and their changes through time.The model combines a dynamic topic model on Gaussian Markov random fields with a logistic stick-breaking process that realizes Dirichlet process. In the experiments, we evaluated the proposed model in terms of interpretability, accuracy in estimating the number of senses, and tracking their changes using both artificial data and real data.We quantitatively verified that the model behaves as expected through evaluation using artificial data.Using the CCOHA corpus, we showed that our model outperforms the baseline model and investigated the semantic changes of several well-known target words.

null

10.18653/v1/2022.emnlp-main.104

null

27,234

inproceedings

gu-etal-2022-learning

Learning Instructions with Unlabeled Data for Zero-Shot Cross-Task Generalization

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.105/

Gu, Yuxian and Ke, Pei and Zhu, Xiaoyan and Huang, Minlie

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1617--1634

Training language models to learn from human instructions for zero-shot cross-task generalization has attracted much attention in NLP communities. Recently, instruction tuning (IT), which fine-tunes a pre-trained language model on a massive collection of tasks described via human-craft instructions, has been shown effective in instruction learning for unseen tasks. However, IT relies on a large amount of human-annotated samples, which restricts its generalization. Unlike labeled data, unlabeled data are often massive and cheap to obtain. In this work, we study how IT can be improved with unlabeled data. We first empirically explore the IT performance trends versus the number of labeled data, instructions, and training tasks. We find it critical to enlarge the number of training instructions, and the instructions can be underutilized due to the scarcity of labeled data. Then, we propose Unlabeled Data Augmented Instruction Tuning (UDIT) to take better advantage of the instructions during IT by constructing pseudo-labeled data from unlabeled plain texts. We conduct extensive experiments to show UDIT`s effectiveness in various scenarios of tasks and datasets. We also comprehensively analyze the key factors of UDIT to investigate how to better improve IT with unlabeled data. The code is publicly available at https://github.com/thu-coai/UDIT.

null

10.18653/v1/2022.emnlp-main.105

null

27,235

inproceedings

ou-etal-2022-counterfactual

Counterfactual Data Augmentation via Perspective Transition for Open-Domain Dialogues

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.106/

Ou, Jiao and Zhang, Jinchao and Feng, Yang and Zhou, Jie

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1635--1648

The construction of open-domain dialogue systems requires high-quality dialogue datasets. The dialogue data admits a wide variety of responses for a given dialogue history, especially responses with different semantics. However, collecting high-quality such a dataset in most scenarios is labor-intensive and time-consuming. In this paper, we propose a data augmentation method to automatically augment high-quality responses with different semantics by counterfactual inference. Specifically, given an observed dialogue, our counterfactual generation model first infers semantically different responses by replacing the observed reply perspective with substituted ones. Furthermore, our data selection method filters out detrimental augmented responses. Experimental results show that our data augmentation method can augment high-quality responses with different semantics for a given dialogue history, and can outperform competitive baselines on multiple downstream tasks.

null

10.18653/v1/2022.emnlp-main.106

null

27,236

inproceedings

bai-etal-2022-squire

{SQUIRE}: A Sequence-to-sequence Framework for Multi-hop Knowledge Graph Reasoning

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.107/

Bai, Yushi and Lv, Xin and Li, Juanzi and Hou, Lei and Qu, Yincen and Dai, Zelin and Xiong, Feiyu

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1649--1662

Multi-hop knowledge graph (KG) reasoning has been widely studied in recent years to provide interpretable predictions on missing links with evidential paths. Most previous works use reinforcement learning (RL) based methods that learn to navigate the path towards the target entity. However, these methods suffer from slow and poor convergence, and they may fail to infer a certain path when there is a missing edge along the path. Here we present SQUIRE, the first Sequence-to-sequence based multi-hop reasoning framework, which utilizes an encoder-decoder Transformer structure to translate the query to a path. Our framework brings about two benefits: (1) It can learn and predict in an end-to-end fashion, which gives better and faster convergence; (2) Our transformer model does not rely on existing edges to generate the path, and has the flexibility to complete missing edges along the path, especially in sparse KGs. Experiments on standard and sparse KGs show that our approach yields significant improvement over prior methods, while converging 4x-7x faster.

null

10.18653/v1/2022.emnlp-main.107

null

27,237

inproceedings

zhang-etal-2022-speechut

{S}peech{UT}: Bridging Speech and Text with Hidden-Unit for Encoder-Decoder Based Speech-Text Pre-training

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.108/

Zhang, Ziqiang and Zhou, Long and Ao, Junyi and Liu, Shujie and Dai, Lirong and Li, Jinyu and Wei, Furu

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1663--1676

The rapid development of single-modal pre-training has prompted researchers to pay more attention to cross-modal pre-training methods. In this paper, we propose a unified-modal speech-unit-text pre-training model, SpeechUT, to connect the representations of a speech encoder and a text decoder with a shared unit encoder. Leveraging hidden-unit as an interface to align speech and text, we can decompose the speech-to-text model into a speech-to-unit model and a unit-to-text model, which can be jointly pre-trained with unpaired speech and text data respectively. Our proposed SpeechUT is fine-tuned and evaluated on automatic speech recognition (ASR) and speech translation (ST) tasks. Experimental results show that SpeechUT gets substantial improvements over strong baselines, and achieves state-of-the-art performance on both the LibriSpeech ASR and MuST-C ST tasks. To better understand the proposed SpeechUT, detailed analyses are conducted. The code and pre-trained models are available at https://aka.ms/SpeechUT.

null

10.18653/v1/2022.emnlp-main.108

null

27,238

inproceedings

chen-etal-2022-learning-label

Learning Label Modular Prompts for Text Classification in the Wild

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.109/

Chen, Hailin and Saha, Amrita and Joty, Shafiq and Hoi, Steven C.H.

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1677--1690

Machine learning models usually assume i.i.d data during training and testing, but data and tasks in real world often change over time. To emulate the transient nature of real world, we propose a challenging but practical task: text classification in-the-wild, which introduces different non-stationary training/testing stages. Decomposing a complex task into modular components can enable robust generalisation under such non-stationary environment. However, current modular approaches in NLP do not take advantage of recent advances in parameter efficient tuning of pretrained language models. To close this gap, we propose ModularPrompt, a label-modular prompt tuning framework for text classification tasks. In ModularPrompt, the input prompt consists of a sequence of soft label prompts, each encoding modular knowledge related to the corresponding class label. In two of most formidable settings, ModularPrompt outperforms relevant baselines by a large margin demonstrating strong generalisation ability. We also conduct comprehensive analysis to validate whether the learned prompts satisfy properties of a modular representation.

null

10.18653/v1/2022.emnlp-main.109

null

27,239

inproceedings

stanojevic-2022-unbiased

Unbiased and Efficient Sampling of Dependency Trees

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.110/

Stanojevi{\'c}, Milo{\v{s}}

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1691--1706

Most computational models of dependency syntax consist of distributions over spanning trees. However, the majority of dependency treebanks require that every valid dependency tree has a single edge coming out of the ROOT node, a constraint that is not part of the definition of spanning trees. For this reason all standard inference algorithms for spanning trees are sub-optimal for inference over dependency trees.Zmigrod et al (2021) proposed algorithms for sampling with and without replacement from the dependency tree distribution that incorporate the single-root constraint. In this paper we show that their fastest algorithm for sampling with replacement, Wilson-RC, is in fact producing biased samples and we provide two alternatives that are unbiased. Additionally, we propose two algorithms (one incremental, one parallel) that reduce the asymptotic runtime of algorithm for sampling k trees without replacement to O(kn{\textasciicircum}3). These algorithms are both asymptotically and practically more efficient.

null

10.18653/v1/2022.emnlp-main.110

null

27,240

inproceedings

gu-etal-2022-continual

Continual Learning of Neural Machine Translation within Low Forgetting Risk Regions

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.111/

Gu, Shuhao and Hu, Bojie and Feng, Yang

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1707--1718

This paper considers continual learning of large-scale pretrained neural machine translation model without accessing the previous training data or introducing model separation. We argue that the widely used regularization-based methods, which perform multi-objective learning with an auxiliary loss, suffer from the misestimate problem and cannot always achieve a good balance between the previous and new tasks. To solve the problem, we propose a two-stage training method based on the local features of the real loss. We first search low forgetting risk regions, where the model can retain the performance on the previous task as the parameters are updated, to avoid the catastrophic forgetting problem. Then we can continually train the model within this region only with the new training data to fit the new task. Specifically, we propose two methods to search the low forgetting risk regions, which are based on the curvature of loss and the impacts of the parameters on the model output, respectively. We conduct experiments on domain adaptation and more challenging language adaptation tasks, and the experimental results show that our method can achieve significant improvements compared with several strong baselines.

null

10.18653/v1/2022.emnlp-main.111

null

27,241

inproceedings

shen-etal-2022-cost

{COST}-{EFF}: Collaborative Optimization of Spatial and Temporal Efficiency with Slenderized Multi-exit Language Models

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.112/

Shen, Bowen and Lin, Zheng and Liu, Yuanxin and Liu, Zhengxiao and Wang, Lei and Wang, Weiping

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1719--1730

Transformer-based pre-trained language models (PLMs) mostly suffer from excessive overhead despite their advanced capacity. For resource-constrained devices, there is an urgent need for a spatially and temporally efficient model which retains the major capacity of PLMs. However, existing statically compressed models are unaware of the diverse complexities between input instances, potentially resulting in redundancy and inadequacy for simple and complex inputs. Also, miniature models with early exiting encounter challenges in the trade-off between making predictions and serving the deeper layers. Motivated by such considerations, we propose a collaborative optimization for PLMs that integrates static model compression and dynamic inference acceleration. Specifically, the PLM is slenderized in width while the depth remains intact, complementing layer-wise early exiting to speed up inference dynamically. To address the trade-off of early exiting, we propose a joint training approach that calibrates slenderization and preserves contributive structures to each exit instead of only the final layer. Experiments are conducted on GLUE benchmark and the results verify the Pareto optimality of our approach at high compression and acceleration rate with 1/8 parameters and 1/19 FLOPs of BERT.

null

10.18653/v1/2022.emnlp-main.112

null

27,242

inproceedings

wan-etal-2022-rescue

Rescue Implicit and Long-tail Cases: Nearest Neighbor Relation Extraction

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.113/

Wan, Zhen and Liu, Qianying and Mao, Zhuoyuan and Cheng, Fei and Kurohashi, Sadao and Li, Jiwei

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1731--1738

Relation extraction (RE) has achieved remarkable progress with the help of pre-trained language models. However, existing RE models are usually incapable of handling two situations: implicit expressions and long-tail relation types, caused by language complexity and data sparsity. In this paper, we introduce a simple enhancement of RE using $k$ nearest neighbors ($k$NN-RE). $k$NN-RE allows the model to consult training relations at test time through a nearest-neighbor search and provides a simple yet effective means to tackle the two issues above. Additionally, we observe that $k$NN-RE serves as an effective way to leverage distant supervision (DS) data for RE. Experimental results show that the proposed $k$NN-RE achieves state-of-the-art performances on a variety of supervised RE datasets, i.e., ACE05, SciERC, and Wiki80, along with outperforming the best model to date on the i2b2 and Wiki80 datasets in the setting of allowing using DS. Our code and models are available at: https://github.com/YukinoWan/kNN-RE.

null

10.18653/v1/2022.emnlp-main.113

null

27,243

inproceedings

chen-etal-2022-storyer

{S}tory{ER}: Automatic Story Evaluation via Ranking, Rating and Reasoning

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.114/

Chen, Hong and Vo, Duc and Takamura, Hiroya and Miyao, Yusuke and Nakayama, Hideki

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1739--1753

Existing automatic story evaluation methods place a premium on story lexical level coherence, deviating from human preference.We go beyond this limitation by considering a novel Story Evaluation method that mimics human preference when judging a story, namely StoryER, which consists of three sub-tasks: Ranking, Rating and Reasoning.Given either a machine-generated or a human-written story, StoryER requires the machine to output 1) a preference score that corresponds to human preference, 2) specific ratings and their corresponding confidences and 3) comments for various aspects (e.g., opening, character-shaping).To support these tasks, we introduce a well-annotated dataset comprising (i) 100k ranked story pairs; and (ii) a set of 46k ratings and comments on various aspects of the story.We finetune Longformer-Encoder-Decoder (LED) on the collected dataset, with the encoder responsible for preference score and aspect prediction and the decoder for comment generation.Our comprehensive experiments result a competitive benchmark for each task, showing the high correlation to human preference.In addition, we have witnessed the joint learning of the preference scores, the aspect ratings, and the comments brings gain each single task.Our dataset and benchmarks are publicly available to advance the research of story evaluation tasks.

null

10.18653/v1/2022.emnlp-main.114

null

27,244

inproceedings

mitchell-etal-2022-enhancing

Enhancing Self-Consistency and Performance of Pre-Trained Language Models through Natural Language Inference

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.115/

Mitchell, Eric and Noh, Joseph and Li, Siyan and Armstrong, Will and Agarwal, Ananth and Liu, Patrick and Finn, Chelsea and Manning, Christopher

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1754--1768

While large pre-trained language models are powerful, their predictions often lack logical consistency across test inputs. For example, a state-of-the-art Macaw question-answering (QA) model answers {\ensuremath{<}}i{\ensuremath{>}}Yes{\ensuremath{<}}/i{\ensuremath{>}} to {\ensuremath{<}}i{\ensuremath{>}}Is a sparrow a bird?{\ensuremath{<}}/i{\ensuremath{>}} and {\ensuremath{<}}i{\ensuremath{>}}Does a bird have feet?{\ensuremath{<}}/i{\ensuremath{>}} but answers {\ensuremath{<}}i{\ensuremath{>}}No{\ensuremath{<}}/i{\ensuremath{>}} to {\ensuremath{<}}i{\ensuremath{>}}Does a sparrow have feet?{\ensuremath{<}}/i{\ensuremath{>}}. To address this failure mode, we propose a framework, Consistency Correction through Relation Detection, or {\ensuremath{<}}b{\ensuremath{>}}ConCoRD{\ensuremath{<}}/b{\ensuremath{>}}, for boosting the consistency and accuracy of pre-trained NLP models using pre-trained natural language inference (NLI) models without fine-tuning or re-training. Given a batch of test inputs, ConCoRD samples several candidate outputs for each input and instantiates a factor graph that accounts for both the model`s belief about the likelihood of each answer choice in isolation and the NLI model`s beliefs about pair-wise answer choice compatibility. We show that a weighted MaxSAT solver can efficiently compute high-quality answer choices under this factor graph, improving over the raw model`s predictions. Our experiments demonstrate that ConCoRD consistently boosts accuracy and consistency of off-the-shelf closed-book QA and VQA models using off-the-shelf NLI models, notably increasing accuracy of LXMERT on ConVQA by 5{\%} absolute. See the project website (https://ericmitchell.ai/emnlp-2022-concord/) for code and data.

null

10.18653/v1/2022.emnlp-main.115

null

27,245

inproceedings

zhang-etal-2022-robustness

Robustness of Demonstration-based Learning Under Limited Data Scenario

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.116/

Zhang, Hongxin and Zhang, Yanzhe and Zhang, Ruiyi and Yang, Diyi

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1769--1782

Demonstration-based learning has shown great potential in stimulating pretrained language models' ability under limited data scenario. Simply augmenting the input with some demonstrations can significantly improve performance on few-shot NER. However, why such demonstrations are beneficial for the learning process remains unclear since there is no explicit alignment between the demonstrations and the predictions. In this paper, we design pathological demonstrations by gradually removing intuitively useful information from the standard ones to take a deep dive of the robustness of demonstration-based sequence labeling and show that (1) demonstrations composed of random tokens still make the model a better few-shot learner; (2) the length of random demonstrations and the relevance of random tokens are the main factors affecting the performance; (3) demonstrations increase the confidence of model predictions on captured superficial patterns. We have publicly released our code at https://github.com/SALT-NLP/RobustDemo.

null

10.18653/v1/2022.emnlp-main.116

null

27,246

inproceedings

wright-etal-2022-modeling

Modeling Information Change in Science Communication with Semantically Matched Paraphrases

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.117/

Wright, Dustin and Pei, Jiaxin and Jurgens, David and Augenstein, Isabelle

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1783--1807

Whether the media faithfully communicate scientific information has long been a core issue to the science community. Automatically identifying paraphrased scientific findings could enable large-scale tracking and analysis of information changes in the science communication process, but this requires systems to understand the similarity between scientific information across multiple domains. To this end, we present the SCIENTIFIC PARAPHRASE AND INFORMATION CHANGE DATASET (SPICED), the first paraphrase dataset of scientific findings annotated for degree of information change. SPICED contains 6,000 scientific finding pairs extracted from news stories, social media discussions, and full texts of original papers. We demonstrate that SPICED poses a challenging task and that models trained on SPICED improve downstream performance on evidence retrieval for fact checking of real-world scientific claims. Finally, we show that models trained on SPICED can reveal large-scale trends in the degrees to which people and organizations faithfully communicate new scientific findings. Data, code, and pre-trained models are available at http://www.copenlu.com/publication/2022{\_}emnlp{\_}wright/.

null

10.18653/v1/2022.emnlp-main.117

null

27,247

inproceedings

lasri-etal-2022-word

Word Order Matters When You Increase Masking

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.118/

Lasri, Karim and Lenci, Alessandro and Poibeau, Thierry

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1808--1815

Word order, an essential property of natural languages, is injected in Transformer-based neural language models using position encoding. However, recent experiments have shown that explicit position encoding is not always useful, since some models without such feature managed to achieve state-of-the art performance on some tasks. To understand better this phenomenon, we examine the effect of removing position encodings on the pre-training objective itself (i.e., masked language modelling), to test whether models can reconstruct position information from co-occurrences alone. We do so by controlling the amount of masked tokens in the input sentence, as a proxy to affect the importance of position information for the task. We find that the necessity of position information increases with the amount of masking, and that masked language models without position encodings are not able to reconstruct this information on the task. These findings point towards a direct relationship between the amount of masking and the ability of Transformers to capture order-sensitive aspects of language using position encoding.

null

10.18653/v1/2022.emnlp-main.118

null

27,248

inproceedings

mireshghallah-etal-2022-empirical

An Empirical Analysis of Memorization in Fine-tuned Autoregressive Language Models

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.119/

Mireshghallah, Fatemehsadat and Uniyal, Archit and Wang, Tianhao and Evans, David and Berg-Kirkpatrick, Taylor

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1816--1826

Large language models are shown to present privacy risks through memorization of training data, andseveral recent works have studied such risks for the pre-training phase. Little attention, however, has been given to the fine-tuning phase and it is not well understood how different fine-tuning methods (such as fine-tuning the full model, the model head, and adapter) compare in terms of memorization risk. This presents increasing concern as the {\textquotedblleft}pre-train and fine-tune{\textquotedblright} paradigm proliferates. In this paper, we empirically study memorization of fine-tuning methods using membership inference and extraction attacks, and show that their susceptibility to attacks is very different. We observe that fine-tuning the head of the model has the highest susceptibility to attacks, whereas fine-tuning smaller adapters appears to be less vulnerable to known extraction attacks.

null

10.18653/v1/2022.emnlp-main.119

null

27,249

inproceedings

chen-etal-2022-style

Style Transfer as Data Augmentation: A Case Study on Named Entity Recognition

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.120/

Chen, Shuguang and Neves, Leonardo and Solorio, Thamar

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1827--1841

In this work, we take the named entity recognition task in the English language as a case study and explore style transfer as a data augmentation method to increase the size and diversity of training data in low-resource scenarios. We propose a new method to effectively transform the text from a high-resource domain to a low-resource domain by changing its style-related attributes to generate synthetic data for training. Moreover, we design a constrained decoding algorithm along with a set of key ingredients for data selection to guarantee the generation of valid and coherent data. Experiments and analysis on five different domain pairs under different data regimes demonstrate that our approach can significantly improve results compared to current state-of-the-art data augmentation methods. Our approach is a practical solution to data scarcity, and we expect it to be applicable to other NLP tasks.

null

10.18653/v1/2022.emnlp-main.120

null

27,250

inproceedings

cardon-etal-2022-linguistic

Linguistic Corpus Annotation for Automatic Text Simplification Evaluation

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.121/

Cardon, R{\'emi and Bibal, Adrien and Wilkens, Rodrigo and Alfter, David and Norr{\'e, Magali and M{\"uller, Adeline and Patrick, Watrin and Fran{\c{cois, Thomas

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1842--1866

Evaluating automatic text simplification (ATS) systems is a difficult task that is either performed by automatic metrics or user-based evaluations. However, from a linguistic point-of-view, it is not always clear on what bases these evaluations operate. In this paper, we propose annotations of the ASSET corpus that can be used to shed more light on ATS evaluation. In addition to contributing with this resource, we show how it can be used to analyze SARI`s behavior and to re-evaluate existing ATS systems. We present our insights as a step to improve ATS evaluation protocols in the future.

null

10.18653/v1/2022.emnlp-main.121

null

27,251

inproceedings

ding-etal-2022-semantic

Semantic Framework based Query Generation for Temporal Question Answering over Knowledge Graphs

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.122/

Ding, Wentao and Chen, Hao and Li, Huayu and Qu, Yuzhong

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1867--1877

Answering factual questions with temporal intent over knowledge graphs (temporal KGQA) attracts rising attention in recent years.In the generation of temporal queries, existing KGQA methods ignore the fact that some intrinsic connections between events can make them temporally related, which may limit their capability.We systematically analyze the possible interpretation of temporal constraints and conclude the interpretation structures as the Semantic Framework of Temporal Constraints, SF-TCons.Based on the semantic framework, we propose a temporal question answering method, SF-TQA, which generates query graphs by exploring the relevant facts of mentioned entities, where the exploring process is restricted by SF-TCons. Our evaluations show that SF-TQA significantly outperforms existing methods on two benchmarks over different knowledge graphs.

null

10.18653/v1/2022.emnlp-main.122

null

27,252

inproceedings

zhao-etal-2022-standard

There Is No Standard Answer: Knowledge-Grounded Dialogue Generation with Adversarial Activated Multi-Reference Learning

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.123/

Zhao, Xueliang and Fu, Tingchen and Tao, Chongyang and Yan, Rui

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1878--1891

Knowledge-grounded dialogue (KGC) shows excellent potential to deliver an engaging and informative response. However, existing approaches emphasize selecting one golden knowledge given a particular dialogue context, overlooking the one-to-many phenomenon in dialogue. As a result, existing paradigm limits the diversity of knowledge selection and generation. To this end, we establish a multi-reference KGC dataset and propose a series of metrics to systematically assess the one-to-many efficacy of existing KGC models. Furthermore, to extend the hypothesis space of knowledge selection to enhance the mapping relationship between multiple knowledge and multiple responses, we devise a span-based variational model and optimize the model in a wake-sleep style with an ameliorated evidence lower bound objective to learn the one-to-many generalization. Both automatic and human evaluations demonstrate the efficacy of our approach.

null

10.18653/v1/2022.emnlp-main.123

null

27,253

inproceedings

baan-etal-2022-stop

Stop Measuring Calibration When Humans Disagree

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.124/

Baan, Joris and Aziz, Wilker and Plank, Barbara and Fernandez, Raquel

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1892--1915

Calibration is a popular framework to evaluate whether a classifier knows when it does not know - i.e., its predictive probabilities are a good indication of how likely a prediction is to be correct. Correctness is commonly estimated against the human majority class. Recently, calibration to human majority has been measured on tasks where humans inherently disagree about which class applies. We show that measuring calibration to human majority given inherent disagreements is theoretically problematic, demonstrate this empirically on the ChaosNLI dataset, and derive several instance-level measures of calibration that capture key statistical properties of human judgements - including class frequency, ranking and entropy.

null

10.18653/v1/2022.emnlp-main.124

null

27,254

inproceedings

nourbakhsh-etal-2022-improving

Improving compositional generalization for multi-step quantitative reasoning in question answering

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.125/

Nourbakhsh, Armineh and Jiao, Cathy and Shah, Sameena and Ros{\'e}, Carolyn

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1916--1932

Quantitative reasoning is an important aspect of question answering, especially when numeric and verbal cues interact to indicate sophisticated, multi-step programs. In this paper, we demonstrate how modeling the compositional nature of quantitative text can enhance the performance and robustness of QA models, allowing them to capture arithmetic logic that is expressed verbally. Borrowing from the literature on semantic parsing, we propose a method that encourages the QA models to adjust their attention patterns and capture input/output alignments that are meaningful to the reasoning task. We show how this strategy improves program accuracy and renders the models more robust against overfitting as the number of reasoning steps grows. Our approach is designed as a standalone module which can be prepended to many existing models and trained in an end-to-end fashion without the need for additional supervisory signal. As part of this exercise, we also create a unified dataset building on four previously released numerical QA datasets over tabular data.

null

10.18653/v1/2022.emnlp-main.125

null

27,255

inproceedings

wiemerslage-etal-2022-comprehensive

A Comprehensive Comparison of Neural Networks as Cognitive Models of Inflection

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.126/

Wiemerslage, Adam and Dudy, Shiran and Kann, Katharina

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1933--1945

Neural networks have long been at the center of a debate around the cognitive mechanism by which humans process inflectional morphology. This debate has gravitated into NLP by way of the question: Are neural networks a feasible account for human behavior in morphological inflection?We address that question by measuring the correlation between human judgments and neural network probabilities for unknown word inflections. We test a larger range of architectures than previously studied on two important tasks for the cognitive processing debate: English past tense, and German number inflection. We find evidence that the Transformer may be a better account of human behavior than LSTMs on these datasets, and that LSTM features known to increase inflection accuracy do not always result in more human-like behavior.

null

10.18653/v1/2022.emnlp-main.126

null

27,256

inproceedings

pramanick-sarkar-2022-visual

Can Visual Context Improve Automatic Speech Recognition for an Embodied Agent?

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.127/

Pramanick, Pradip and Sarkar, Chayan

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1946--1957

The usage of automatic speech recognition (ASR) systems are becoming omnipresent ranging from personal assistant to chatbots, home, and industrial automation systems, etc. Modern robots are also equipped with ASR capabilities for interacting with humans as speech is the most natural interaction modality. However, ASR in robots faces additional challenges as compared to a personal assistant. Being an embodied agent, a robot must recognize the physical entities around it and therefore reliably recognize the speech containing the description of such entities. However, current ASR systems are often unable to do so due to limitations in ASR training, such as generic datasets and open-vocabulary modeling. Also, adverse conditions during inference, such as noise, accented, and far-field speech makes the transcription inaccurate. In this work, we present a method to incorporate a robot`s visual information into an ASR system and improve the recognition of a spoken utterance containing a visible entity. Specifically, we propose a new decoder biasing technique to incorporate the visual context while ensuring the ASR output does not degrade for incorrect context. We achieve a 59{\%} relative reduction in WER from an unmodified ASR system.

null

10.18653/v1/2022.emnlp-main.127

null

27,257

inproceedings

adebara-etal-2022-afrolid

{A}fro{LID}: A Neural Language Identification Tool for {A}frican Languages

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.128/

Adebara, Ife and Elmadany, AbdelRahim and Abdul-Mageed, Muhammad and Inciarte, Alcides

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1958--1981

Language identification (LID) is a crucial precursor for NLP, especially for mining web data. Problematically, most of the world`s 7000+ languages today are not covered by LID technologies. We address this pressing issue for Africa by introducing AfroLID, a neural LID toolkit for 517 African languages and varieties. AfroLID exploits a multi-domain web dataset manually curated from across 14 language families utilizing five orthographic systems. When evaluated on our blind Test set, AfroLID achieves 95.89 F{\_}1-score. We also compare AfroLID to five existing LID tools that each cover a small number of African languages, finding it to outperform them on most languages. We further show the utility of AfroLID in the wild by testing it on the acutely under-served Twitter domain. Finally, we offer a number of controlled case studies and perform a linguistically-motivated error analysis that allow us to both showcase AfroLID`s powerful capabilities and limitations

null

10.18653/v1/2022.emnlp-main.128

null

27,258

inproceedings

spiliopoulou-etal-2022-events

{E}v{E}nt{S} {R}ea{LM}: Event Reasoning of Entity States via Language Models

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.129/

Spiliopoulou, Evangelia and Pagnoni, Artidoro and Bisk, Yonatan and Hovy, Eduard

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1982--1997

This paper investigates models of event implications. Specifically, how well models predict entity state-changes, by targeting their understanding of physical attributes. Nominally, Large Language models (LLM) have been exposed to procedural knowledge about how objects interact, yet our benchmarking shows they fail to reason about the world. Conversely, we also demonstrate that existing approaches often misrepresent the surprising abilities of LLMs via improper task encodings and that proper model prompting can dramatically improve performance of reported baseline results across multiple tasks. In particular, our results indicate that our prompting technique is especially useful for unseen attributes (out-of-domain) or when only limited data is available.

null

10.18653/v1/2022.emnlp-main.129

null

27,259

inproceedings

agrawal-etal-2022-large

Large language models are few-shot clinical information extractors

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.130/

Agrawal, Monica and Hegselmann, Stefan and Lang, Hunter and Kim, Yoon and Sontag, David

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

1998--2022

A long-running goal of the clinical NLP community is the extraction of important variables trapped in clinical notes. However, roadblocks have included dataset shift from the general domain and a lack of public clinical corpora and annotations. In this work, we show that large language models, such as InstructGPT (Ouyang et al., 2022), perform well at zero- and few-shot information extraction from clinical text despite not being trained specifically for the clinical domain. Whereas text classification and generation performance have already been studied extensively in such models, here we additionally demonstrate how to leverage them to tackle a diverse set of NLP tasks which require more structured outputs, including span identification, token-level sequence classification, and relation extraction. Further, due to the dearth of available data to evaluate these systems, we introduce new datasets for benchmarking few-shot clinical information extraction based on a manual re-annotation of the CASI dataset (Moon et al., 2014) for new tasks. On the clinical extraction tasks we studied, the GPT-3 systems significantly outperform existing zero- and few-shot baselines.

null

10.18653/v1/2022.emnlp-main.130

null

27,260

inproceedings

zhong-etal-2022-towards

Towards a Unified Multi-Dimensional Evaluator for Text Generation

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.131/

Zhong, Ming and Liu, Yang and Yin, Da and Mao, Yuning and Jiao, Yizhu and Liu, Pengfei and Zhu, Chenguang and Ji, Heng and Han, Jiawei

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2023--2038

Multi-dimensional evaluation is the dominant paradigm for human evaluation in Natural Language Generation (NLG), i.e., evaluating the generated text from multiple explainable dimensions, such as coherence and fluency. However, automatic evaluation in NLG is still dominated by similarity-based metrics, and we lack a reliable framework for a more comprehensive evaluation of advanced models. In this paper, we propose a unified multi-dimensional evaluator UniEval for NLG. We re-frame NLG evaluation as a Boolean Question Answering (QA) task, and by guiding the model with different questions, we can use one evaluator to evaluate from multiple dimensions. Furthermore, thanks to the unified Boolean QA format, we are able to introduce an intermediate learning phase that enables UniEval to incorporate external knowledge from multiple related tasks and gain further improvement. Experiments on three typical NLG tasks show that UniEval correlates substantially better with human judgments than existing metrics. Specifically, compared to the top-performing unified evaluators, UniEval achieves a 23{\%} higher correlation on text summarization, and over 43{\%} on dialogue response generation. Also, UniEval demonstrates a strong zero-shot learning ability for unseen evaluation dimensions and tasks. Source code, data, and all pre-trained evaluators are available at https://github.com/maszhongming/UniEval.

null

10.18653/v1/2022.emnlp-main.131

null

27,261

inproceedings

yin-etal-2022-geomlama

{G}eo{MLAMA}: Geo-Diverse Commonsense Probing on Multilingual Pre-Trained Language Models

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.132/

Yin, Da and Bansal, Hritik and Monajatipoor, Masoud and Li, Liunian Harold and Chang, Kai-Wei

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2039--2055

Recent work has shown that Pre-trained Language Models (PLMs) store the relational knowledge learned from data and utilize it for performing downstream tasks. However, commonsense knowledge across different regions may vary. For instance, the color of bridal dress is white in American weddings whereas it is red in Chinese weddings. In this paper, we introduce a benchmark dataset, Geo-diverse Commonsense Multilingual Language Models Analysis (GeoMLAMA), for probing the diversity of the relational knowledge in multilingual PLMs. GeoMLAMA contains 3125 prompts in English, Chinese, Hindi, Persian, and Swahili, with a wide coverage of concepts shared by people from American, Chinese, Indian, Iranian and Kenyan cultures. We benchmark 11 standard multilingual PLMs on GeoMLAMA. Interestingly, we find that 1) larger multilingual PLMs variants do not necessarily store geo-diverse concepts better than its smaller variant; 2) multilingual PLMs are not intrinsically biased towards knowledge from the Western countries (the United States); 3) the native language of a country may not be the best language to probe its knowledge and 4) a language may better probe knowledge about a non-native country than its native country.

null

10.18653/v1/2022.emnlp-main.132

null

27,262

inproceedings

espana-bonet-barron-cedeno-2022-undesired

The (Undesired) Attenuation of Human Biases by Multilinguality

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.133/

Espa{\~n}a-Bonet, Cristina and Barr{\'o}n-Cede{\~n}o, Alberto

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2056--2077

Some human preferences are universal. The odor of vanilla is perceived as pleasant all around the world. We expect neural models trained on human texts to exhibit these kind of preferences, i.e. biases, but we show that this is not always the case. We explore 16 static and contextual embedding models in 9 languages and, when possible, compare them under similar training conditions. We introduce and release CA-WEAT, multilingual cultural aware tests to quantify biases, and compare them to previous English-centric tests. Our experiments confirm that monolingual static embeddings do exhibit human biases, but values differ across languages, being far from universal. Biases are less evident in contextual models, to the point that the original human association might be reversed. Multilinguality proves to be another variable that attenuates and even reverses the effect of the bias, specially in contextual multilingual models. In order to explain this variance among models and languages, we examine the effect of asymmetries in the training corpus, departures from isomorphism in multilingual embedding spaces and discrepancies in the testing measures between languages.

null

10.18653/v1/2022.emnlp-main.133

null

27,263

inproceedings

tafjord-etal-2022-entailer

Entailer: Answering Questions with Faithful and Truthful Chains of Reasoning

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.134/

Tafjord, Oyvind and Dalvi Mishra, Bhavana and Clark, Peter

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2078--2093

Our goal is a question-answering (QA) system that can show how its answers are implied by its own internal beliefs via a systematic chain of reasoning. Such a capability would allow better understanding of why a model produced the answer it did. Our approach is to recursively combine a trained backward-chainingmodel, capable of generating a set of premises entailing an answer hypothesis, with a verifier that checks that the model itself believes those premises (and the entailment itself) through self-querying. To our knowledge, this is the first system to generate multistep chains that are both faithful (the answer follows from the reasoning) and truthful (the chain reflects the system`s own internal beliefs). In evaluation using two different datasets, users judge that a majority (70{\%}+) of generated chains clearly show how an answer follows from a set of facts - substantially better than a high-performance baseline - while preserving answer accuracy. By materializing model beliefs that systematically support an answer, new opportunities arise for understanding the model`s system of belief, and diagnosing and correcting its misunderstandings when an answer is wrong.

null

10.18653/v1/2022.emnlp-main.134

null

27,264

inproceedings

laban-etal-2022-near

Near-Negative Distinction: Giving a Second Life to Human Evaluation Datasets

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.135/

Laban, Philippe and Wu, Chien-Sheng and Liu, Wenhao and Xiong, Caiming

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2094--2108

Precisely assessing the progress in natural language generation (NLG) tasks is challenging, and human evaluation to establish a preference in a model`s output over another is often necessary.However, human evaluation is usually costly, difficult to reproduce, and non-reusable.In this paper, we propose a new and simple automatic evaluation method for NLG called Near-Negative Distinction (NND) that repurposes prior human annotations into NND tests.In an NND test, an NLG model must place a higher likelihood on a high-quality output candidate than on a near-negative candidate with a known error.Model performance is established by the number of NND tests a model passes, as well as the distribution over task-specific errors the model fails on.Through experiments on three NLG tasks (question generation, question answering, and summarization), we show that NND achieves a higher correlation with human judgments than standard NLG evaluation metrics. We then illustrate NND evaluation in four practical scenarios, for example performing fine-grain model analysis, or studying model training dynamics. Our findings suggest that NND can give a second life to human annotations and provide low-cost NLG evaluation.

null

10.18653/v1/2022.emnlp-main.135

null

27,265

inproceedings

alkhamissi-etal-2022-token

{T}o{K}en: Task Decomposition and Knowledge Infusion for Few-Shot Hate Speech Detection

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.136/

AlKhamissi, Badr and Ladhak, Faisal and Iyer, Srinivasan and Stoyanov, Veselin and Kozareva, Zornitsa and Li, Xian and Fung, Pascale and Mathias, Lambert and Celikyilmaz, Asli and Diab, Mona

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2109--2120

Hate speech detection is complex; it relies on commonsense reasoning, knowledge of stereotypes, and an understanding of social nuance that differs from one culture to the next. It is also difficult to collect a large-scale hate speech annotated dataset. In this work, we frame this problem as a few-shot learning task, and show significant gains with decomposing the task into its {\textquotedblleft}constituent{\textquotedblright} parts. In addition, we see that infusing knowledge from reasoning datasets (e.g. ATOMIC2020) improves the performance even further. Moreover, we observe that the trained models generalize to out-of-distribution datasets, showing the superiority of task decomposition and knowledge infusion compared to previously used methods. Concretely, our method outperforms the baseline by 17.83{\%} absolute gain in the 16-shot case.

null

10.18653/v1/2022.emnlp-main.136

null

27,266

inproceedings

saha-etal-2022-hard

Are Hard Examples also Harder to Explain? A Study with Human and Model-Generated Explanations

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.137/

Saha, Swarnadeep and Hase, Peter and Rajani, Nazneen and Bansal, Mohit

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2121--2131

Recent work on explainable NLP has shown that few-shot prompting can enable large pre-trained language models (LLMs) to generate grammatical and factual natural language explanations for data labels. In this work, we study the connection between explainability and sample hardness by investigating the following research question {--} {\textquotedblleft}Are LLMs and humans equally good at explaining data labels for both easy and hard samples?{\textquotedblright} We answer this question by first collecting human-written explanations in the form of generalizable commonsense rules on the task of Winograd Schema Challenge (Winogrande dataset). We compare these explanations with those generated by GPT-3 while varying the hardness of the test samples as well as the in-context samples. We observe that (1) GPT-3 explanations are as grammatical as human explanations regardless of the hardness of the test samples, (2) for easy examples, GPT-3 generates highly supportive explanations but human explanations are more generalizable, and (3) for hard examples, human explanations are significantly better than GPT-3 explanations both in terms of label-supportiveness and generalizability judgements. We also find that hardness of the in-context examples impacts the quality of GPT-3 explanations. Finally, we show that the supportiveness and generalizability aspects of human explanations are also impacted by sample hardness, although by a much smaller margin than models.

null

10.18653/v1/2022.emnlp-main.137

null

27,267

inproceedings

zheng-etal-2022-stanceosaurus

Stanceosaurus: Classifying Stance Towards Multicultural Misinformation

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.138/

Zheng, Jonathan and Baheti, Ashutosh and Naous, Tarek and Xu, Wei and Ritter, Alan

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2132--2151

We present Stanceosaurus, a new corpus of 28,033 tweets in English, Hindi and Arabic annotated with stance towards 250 misinformation claims. As far as we are aware, it is the largest corpus annotated with stance towards misinformation claims. The claims in Stanceosaurus originate from 15 fact-checking sources that cover diverse geographical regions and cultures. Unlike existing stance datasets, we introduce a more fine-grained 5-class labeling strategy with additional subcategories to distinguish implicit stance. Pre-trained transformer-based stance classifiers that are fine-tuned on our corpus show good generalization on unseen claims and regional claims from countries outside the training data. Cross-lingual experiments demonstrate Stanceosaurus' capability of training multilingual models, achieving 53.1 F1 on Hindi and 50.4 F1 on Arabic without any target-language fine-tuning. Finally, we show how a domain adaptation method can be used to improve performance on Stanceosaurus using additional RumourEval-2019 data. We will make Stanceosaurus publicly available to the research community upon publication and hope it will encourage further work on misinformation identification across languages and cultures.

null

10.18653/v1/2022.emnlp-main.138

null

27,268

inproceedings

lin-etal-2022-gendered

Gendered Mental Health Stigma in Masked Language Models

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.139/

Lin, Inna and Njoo, Lucille and Field, Anjalie and Sharma, Ashish and Reinecke, Katharina and Althoff, Tim and Tsvetkov, Yulia

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2152--2170

Mental health stigma prevents many individuals from receiving the appropriate care, and social psychology studies have shown that mental health tends to be overlooked in men. In this work, we investigate gendered mental health stigma in masked language models. In doing so, we operationalize mental health stigma by developing a framework grounded in psychology research: we use clinical psychology literature to curate prompts, then evaluate the models' propensity to generate gendered words. We find that masked language models capture societal stigma about gender in mental health: models are consistently more likely to predict female subjects than male in sentences about having a mental health condition (32{\%} vs. 19{\%}), and this disparity is exacerbated for sentences that indicate treatment-seeking behavior. Furthermore, we find that different models capture dimensions of stigma differently for men and women, associating stereotypes like anger, blame, and pity more with women with mental health conditions than with men. In showing the complex nuances of models' gendered mental health stigma, we demonstrate that context and overlapping dimensions of identity are important considerations when assessing computational models' social biases.

null

10.18653/v1/2022.emnlp-main.139

null

27,269

inproceedings

yadav-etal-2022-efficient

Efficient Nearest Neighbor Search for Cross-Encoder Models using Matrix Factorization

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.140/

Yadav, Nishant and Monath, Nicholas and Angell, Rico and Zaheer, Manzil and McCallum, Andrew

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2171--2194

Efficient k-nearest neighbor search is a fundamental task, foundational for many problems in NLP. When the similarity is measured by dot-product between dual-encoder vectors or L2-distance, there already exist many scalable and efficient search methods. But not so when similarity is measured by more accurate and expensive black-box neural similarity models, such as cross-encoders, which jointly encode the query and candidate neighbor. The cross-encoders' high computational cost typically limits their use to reranking candidates retrieved by a cheaper model, such as dual encoder or TF-IDF. However, the accuracy of such a two-stage approach is upper-bounded by the recall of the initial candidate set, and potentially requires additional training to align the auxiliary retrieval model with the cross-encoder model. In this paper, we present an approach that avoids the use of a dual-encoder for retrieval, relying solely on the cross-encoder. Retrieval is made efficient with CUR decomposition, a matrix decomposition approach that approximates all pairwise cross-encoder distances from a small subset of rows and columns of the distance matrix. Indexing items using our approach is computationally cheaper than training an auxiliary dual-encoder model through distillation. Empirically, for k {\ensuremath{>}} 10, our approach provides test-time recall-vs-computational cost trade-offs superior to the current widely-used methods that re-rank items retrieved using a dual-encoder or TF-IDF.

null

10.18653/v1/2022.emnlp-main.140

null

27,270

inproceedings

suzgun-etal-2022-prompt

Prompt-and-Rerank: A Method for Zero-Shot and Few-Shot Arbitrary Textual Style Transfer with Small Language Models

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.141/

Suzgun, Mirac and Melas-Kyriazi, Luke and Jurafsky, Dan

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2195--2222

We propose a method for arbitrary textual style transfer (TST){---}the task of transforming a text into any given style{---}utilizing general-purpose pre-trained language models. Our method, Prompt-and-Rerank, is based on a mathematical formulation of the TST task, decomposing it into three constituent components: textual similarity, target style strength, and fluency. Our method uses zero-shot or few-shot prompting to obtain a set of candidate generations in the target style, and then re-ranks them according to the three components. Our method enables small pre-trained language models to perform on par with state-of-the-art large-scale models while using two orders of magnitude less compute and memory. We also investigate the effect of model size and prompt design (e.g., prompt paraphrasing and delimiter-pair choice) on style transfer quality across seven diverse textual style transfer datasets, finding, among other things, that delimiter-pair choice has a large impact on performance, and that models have biases on the direction of style transfer.

null

10.18653/v1/2022.emnlp-main.141

null

27,271

inproceedings

zhou-etal-2022-learning-decompose

Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Texts

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.142/

Zhou, Ben and Richardson, Kyle and Yu, Xiaodong and Roth, Dan

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2223--2235

Explicit decomposition modeling, which involves breaking down complex tasks into more straightforward and often more interpretable sub-tasks, has long been a central theme in developing robust and interpretable NLU systems. However, despite the many datasets and resources built as part of this effort, the majority have small-scale annotations and limited scope, which is insufficient to solve general decomposition tasks. In this paper, we look at large-scale intermediate pre-training of decomposition-based transformers using distant supervision from comparable texts, particularly large-scale parallel news. We show that with such intermediate pre-training, developing robust decomposition-based models for a diverse range of tasks becomes more feasible. For example, on semantic parsing, our model, DecompT5, improves 20{\%} to 30{\%} on two datasets, Overnight and TORQUE, over the baseline language model. We further use DecompT5 to build a novel decomposition-based QA system named DecompEntail, improving over state-of-the-art models, including GPT-3, on both HotpotQA and StrategyQA by 8{\%} and 4{\%}, respectively.

null

10.18653/v1/2022.emnlp-main.142

null

27,272

inproceedings

diwan-etal-2022-winoground

Why is Winoground Hard? Investigating Failures in Visuolinguistic Compositionality

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.143/

Diwan, Anuj and Berry, Layne and Choi, Eunsol and Harwath, David and Mahowald, Kyle

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2236--2250

Recent visuolinguistic pre-trained models show promising progress on various end tasks such as image retrieval and video captioning. Yet, they fail miserably on the recently proposed Winoground dataset, which challenges models to match paired images and English captions, with items constructed to overlap lexically but differ in meaning (e.g., {\textquotedblleft}there is a mug in some grass{\textquotedblright} vs. {\textquotedblleft}there is some grass in a mug{\textquotedblright}). By annotating the dataset using new fine-grained tags, we show that solving the Winoground task requires not just compositional language understanding, but a host of other abilities like commonsense reasoning or locating small, out-of-focus objects in low-resolution images. In this paper, we identify the dataset`s main challenges through a suite of experiments on related tasks (probing task, image retrieval task), data augmentation, and manual inspection of the dataset. Our analysis suggests that a main challenge in visuolinguistic models may lie in fusing visual and textual representations, rather than in compositional language understanding. We release our annotation and code at https://github.com/ajd12342/why-winoground-hard.

null

10.18653/v1/2022.emnlp-main.143

null

27,273

inproceedings

kumar-etal-2022-gradient

Gradient-based Constrained Sampling from Language Models

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.144/

Kumar, Sachin and Paria, Biswajit and Tsvetkov, Yulia

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2251--2277

Large pretrained language models are successful at generating fluent text but are notoriously hard to controllably sample from. In this work, we study constrained sampling from such language models, i.e., generating text that satisfies user-defined constraints, while maintaining fluency and model`s performance in a downstream task. We propose MuCoLa{---}a sampling procedure that combines the log-likelihood of the language model with arbitrary (differentiable) constraints in a single energy function, and then generates samples in a non-autoregressive manner. Specifically, it initializes the entire output sequence with noise and follows a Markov chain defined by Langevin Dynamics using the gradients of this energy. We evaluate MuCoLa on text generation with soft and hard constraints as well as their combinations, obtaining significant improvements over competitive baselines for toxicity avoidance, sentiment control, and keyword-guided generation.

null

10.18653/v1/2022.emnlp-main.144

null

27,274

inproceedings

zhou-etal-2022-tacube

{T}a{C}ube: Pre-computing Data Cubes for Answering Numerical-Reasoning Questions over Tabular Data

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.145/

Zhou, Fan and Hu, Mengkang and Dong, Haoyu and Cheng, Zhoujun and Cheng, Fan and Han, Shi and Zhang, Dongmei

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2278--2291

Existing auto-regressive pre-trained language models (PLMs) like T5 and BART, have been well applied to table question answering by UNIFIEDSKG and TAPEX, respectively, and demonstrated state-of-the-art results on multiple benchmarks. However, auto-regressive PLMs are challenged by recent emerging numerical reasoning datasets, such as TAT-QA, due to the error-prone implicit calculation. In this paper, we present TaCube, to pre-compute aggregation/arithmetic results for the table in advance, so that they are handy and readily available for PLMs to answer numerical reasoning questions. TaCube systematically and comprehensively covers a collection of computational operations over table segments. By simply concatenating TaCube to the input sequence of PLMs, it shows significant experimental effectiveness. TaCube promotes the F1 score from 49.6{\%} to 66.2{\%} on TAT-QA and achieves new state-of-the-art results on WikiTQ (59.6{\%} denotation accuracy). TaCube`s improvements on numerical reasoning cases are even more notable: on TAT-QA, TaCube promotes the exact match accuracy of BART-large by 39.6{\%} on sum, 52.5{\%} on average, 36.6{\%} on substraction, and 22.2{\%} on division. We believe that TaCube is a general and portable pre-computation solution that can be potentially integrated to various numerical reasoning frameworks

null

10.18653/v1/2022.emnlp-main.145

null

27,275

inproceedings

chen-etal-2022-rich

Rich Knowledge Sources Bring Complex Knowledge Conflicts: Recalibrating Models to Reflect Conflicting Evidence

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.146/

Chen, Hung-Ting and Zhang, Michael and Choi, Eunsol

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2292--2307

Question answering models can use rich knowledge sources {---} up to one hundred retrieved passages and parametric knowledge in the large-scale language model (LM). Prior work assumes information in such knowledge sources is consistent with each other, paying little attention to how models blend information stored in their LM parameters with that from retrieved evidence documents. In this paper, we simulate knowledge conflicts (i.e., where parametric knowledge suggests one answer and different passages suggest different answers) and examine model behaviors. We find retrieval performance heavily impacts which sources models rely on, and current models mostly rely on non-parametric knowledgein their best-performing settings. We discover a troubling trend that contradictions among knowledge sources affect model confidence only marginally. To address this issue, we present a new calibration study, where models are discouraged from presenting any single answer when presented with multiple conflicting answer candidates in retrieved evidences.

null

10.18653/v1/2022.emnlp-main.146

null

27,276

inproceedings

yue-etal-2022-qa

{QA} Domain Adaptation using Hidden Space Augmentation and Self-Supervised Contrastive Adaptation

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.147/

Yue, Zhenrui and Zeng, Huimin and Kratzwald, Bernhard and Feuerriegel, Stefan and Wang, Dong

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2308--2321

Question answering (QA) has recently shown impressive results for answering questions from customized domains. Yet, a common challenge is to adapt QA models to an unseen target domain. In this paper, we propose a novel self-supervised framework called QADA for QA domain adaptation. QADA introduces a novel data augmentation pipeline used to augment training QA samples. Different from existing methods, we enrich the samples via hidden space augmentation. For questions, we introduce multi-hop synonyms and sample augmented token embeddings with Dirichlet distributions. For contexts, we develop an augmentation method which learns to drop context spans via a custom attentive sampling strategy. Additionally, contrastive learning is integrated in the proposed self-supervised adaptation framework QADA. Unlike existing approaches, we generate pseudo labels and propose to train the model via a novel attention-based contrastive adaptation method. The attention weights are used to build informative features for discrepancy estimation that helps the QA model separate answers and generalize across source and target domains. To the best of our knowledge, our work is the first to leverage hidden space augmentation and attention-based contrastive adaptation for self-supervised domain adaptation in QA. Our evaluation shows that QADA achieves considerable improvements on multiple target datasets over state-of-the-art baselines in QA domain adaptation.

null

10.18653/v1/2022.emnlp-main.147

null

27,277

inproceedings

shah-etal-2022-flue

When {FLUE} Meets {FLANG}: Benchmarks and Large Pretrained Language Model for Financial Domain

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.148/

Shah, Raj and Chawla, Kunal and Eidnani, Dheeraj and Shah, Agam and Du, Wendi and Chava, Sudheer and Raman, Natraj and Smiley, Charese and Chen, Jiaao and Yang, Diyi

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2322--2335

Pre-trained language models have shown impressive performance on a variety of tasks and domains. Previous research on financial language models usually employs a generic training scheme to train standard model architectures, without completely leveraging the richness of the financial data. We propose a novel domain specific Financial LANGuage model (FLANG) which uses financial keywords and phrases for better masking, together with span boundary objective and in-filing objective. Additionally, the evaluation benchmarks in the field have been limited. To this end, we contribute the Financial Language Understanding Evaluation (FLUE), an open-source comprehensive suite of benchmarks for the financial domain. These include new benchmarks across 5 NLP tasks in financial domain as well as common benchmarks used in the previous research. Experiments on these benchmarks suggest that our model outperforms those in prior literature on a variety of NLP tasks. Our models, code and benchmark data will be made publicly available on Github and Huggingface.

null

10.18653/v1/2022.emnlp-main.148

null

27,278

inproceedings

jiang-etal-2022-retrieval

Retrieval as Attention: End-to-end Learning of Retrieval and Reading within a Single Transformer

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.149/

Jiang, Zhengbao and Gao, Luyu and Wang, Zhiruo and Araki, Jun and Ding, Haibo and Callan, Jamie and Neubig, Graham

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2336--2349

Systems for knowledge-intensive tasks such as open-domain question answering (QA) usually consist of two stages: efficient retrieval of relevant documents from a large corpus and detailed reading of the selected documents. This is usually done through two separate models, a retriever that encodes the query and finds nearest neighbors, and a reader based on Transformers. These two components are usually modeled separately, which necessitates a cumbersome implementation and is awkward to optimize in an end-to-end fashion. In this paper, we revisit this design and eschew the separate architecture and training in favor of a single Transformer that performs retrieval as attention (RAA), and end-to-end training solely based on supervision from the end QA task. We demonstrate for the first time that an end-to-end trained single Transformer can achieve both competitive retrieval and QA performance on in-domain datasets, matching or even slightly outperforming state-of-the-art dense retrievers and readers. Moreover, end-to-end adaptation of our model significantly boosts its performance on out-of-domain datasets in both supervised and unsupervised settings, making our model a simple and adaptable end-to-end solution for knowledge-intensive tasks.

null

10.18653/v1/2022.emnlp-main.149

null

27,279

inproceedings

arvan-etal-2022-reproducibility-computational

Reproducibility in Computational Linguistics: Is Source Code Enough?

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.150/

Arvan, Mohammad and Pina, Lu{\'i}s and Parde, Natalie

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2350--2361

The availability of source code has been put forward as one of the most critical factors for improving the reproducibility of scientific research. This work studies trends in source code availability at major computational linguistics conferences, namely, ACL, EMNLP, LREC, NAACL, and COLING. We observe positive trends, especially in conferences that actively promote reproducibility. We follow this by conducting a reproducibility study of eight papers published in EMNLP 2021, finding that source code releases leave much to be desired. Moving forward, we suggest all conferences require self-contained artifacts and provide a venue to evaluate such artifacts at the time of publication. Authors can include small-scale experiments and explicit scripts to generate each result to improve the reproducibility of their work.

null

10.18653/v1/2022.emnlp-main.150

null

27,280

inproceedings

kim-etal-2022-generating

Generating Information-Seeking Conversations from Unlabeled Documents

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.151/

Kim, Gangwoo and Kim, Sungdong and Yoo, Kang Min and Kang, Jaewoo

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2362--2378

Synthesizing datasets for conversational question answering (CQA) from unlabeled documents remains challenging due to its interactive nature.Moreover, while modeling information needs is an essential key, only few studies have discussed it.In this paper, we introduce a novel framework, **SimSeek**, (**Sim**ulating information-**Seek**ing conversation from unlabeled documents), and compare its two variants.In our baseline, **SimSeek-sym**, a questioner generates follow-up questions upon the predetermined answer by an answerer.On the contrary, **SimSeek-asym** first generates the question and then finds its corresponding answer under the conversational context.Our experiments show that they can synthesize effective training resources for CQA and conversational search tasks.As a result, conversations from **SimSeek-asym** not only make more improvements in our experiments but also are favorably reviewed in a human evaluation.We finally release a large-scale resource of synthetic conversations, **Wiki-SimSeek**, containing 2 million CQA pairs built upon Wikipedia documents.With the dataset, our CQA model achieves the state-of-the-art performance on a recent CQA benchmark, QuAC.The code and dataset are available at https://github.com/naver-ai/simseek

null

10.18653/v1/2022.emnlp-main.151

null

27,281

inproceedings

peng-etal-2022-distill

Distill The Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.152/

Peng, Ru and Zeng, Yawen and Zhao, Jake

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2379--2390

Past works on multimodal machine translation (MMT) elevate bilingual setup by incorporating additional aligned vision information.However, an image-must requirement of the multimodal dataset largely hinders MMT`s development {---} namely that it demands an aligned form of [image, source text, target text].This limitation is generally troublesome during the inference phase especially when the aligned image is not provided as in the normal NMT setup.Thus, in this work, we introduce IKD-MMT, a novel MMT framework to support the image-free inference phase via an inversion knowledge distillation scheme.In particular, a multimodal feature generator is executed with a knowledge distillation module, which directly generates the multimodal feature from (only) source texts as the input.While there have been a few prior works entertaining the possibility to support image-free inference for machine translation, their performances have yet to rival the image-must translation.In our experiments, we identify our method as the first image-free approach to comprehensively rival or even surpass (almost) all image-must frameworks, and achieved the state-of-the-art result on the often-used Multi30k benchmark. Our code and data are availableat: https://github.com/pengr/IKD-mmt/tree/master..

null

10.18653/v1/2022.emnlp-main.152

null

27,282

inproceedings

altakrori-etal-2022-multifaceted

A Multifaceted Framework to Evaluate Evasion, Content Preservation, and Misattribution in Authorship Obfuscation Techniques

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.153/

Altakrori, Malik and Scialom, Thomas and Fung, Benjamin C. M. and Cheung, Jackie Chi Kit

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2391--2406

Authorship obfuscation techniques have commonly been evaluated based on their ability to hide the author`s identity (evasion) while preserving the content of the original text. However, to avoid overstating the systems' effectiveness, evasion detection must be evaluated using competitive identification techniques in settings that mimic real-life scenarios, and the outcomes of the content-preservation evaluation have to be interpretable by potential users of these obfuscation tools. Motivated by recent work on cross-topic authorship identification and content preservation in summarization, we re-evaluate different authorship obfuscation techniques on detection evasion and content preservation. Furthermore, we propose a new information-theoretic measure to characterize the misattribution harm that can be caused by detection evasion. Our results reveal key weaknesses in state-of-the-art obfuscation techniques and a surprisingly competitive effectiveness from a back-translation baseline in all evaluation aspects.

null

10.18653/v1/2022.emnlp-main.153

null

27,283

inproceedings

levy-etal-2022-safetext

{S}afe{T}ext: A Benchmark for Exploring Physical Safety in Language Models

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.154/

Levy, Sharon and Allaway, Emily and Subbiah, Melanie and Chilton, Lydia and Patton, Desmond and McKeown, Kathleen and Wang, William Yang

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2407--2421

Understanding what constitutes safe text is an important issue in natural language processing and can often prevent the deployment of models deemed harmful and unsafe. One such type of safety that has been scarcely studied is commonsense physical safety, i.e. text that is not explicitly violent and requires additional commonsense knowledge to comprehend that it leads to physical harm. We create the first benchmark dataset, SafeText, comprising real-life scenarios with paired safe and physically unsafe pieces of advice. We utilize SafeText to empirically study commonsense physical safety across various models designed for text generation and commonsense reasoning tasks. We find that state-of-the-art large language models are susceptible to the generation of unsafe text and have difficulty rejecting unsafe advice. As a result, we argue for further studies of safety and the assessment of commonsense physical safety in models before release.

null

10.18653/v1/2022.emnlp-main.154

null

27,284

inproceedings

yoo-etal-2022-ground

Ground-Truth Labels Matter: A Deeper Look into Input-Label Demonstrations

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.155/

Yoo, Kang Min and Kim, Junyeob and Kim, Hyuhng Joon and Cho, Hyunsoo and Jo, Hwiyeol and Lee, Sang-Woo and Lee, Sang-goo and Kim, Taeuk

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2422--2437

Despite recent explosion of interests in in-context learning, the underlying mechanism and the precise impact of the quality of demonstrations remain elusive.Intuitively, ground-truth labels should have as much impact in in-context learning (ICL) as supervised learning, but recent work reported that the input-label correspondence is significantly less important than previously thought.Intrigued by this counter-intuitive observation, we re-examine the importance of ground-truth labels in in-context learning.With the introduction of two novel metrics, namely Label-Correctness Sensitivity and Ground-truth Label Effect Ratio (GLER), we were able to conduct quantifiable analysis on the impact of ground-truth label demonstrations.Through extensive analyses, we find that the correct input-label mappings can have varying impacts on the downstream in-context learning performances, depending on the experimental configuration.Through additional studies, we identify key components, such as the verbosity of prompt templates and the language model size, as the controlling factor to achieve more noise-resilient ICL.

null

10.18653/v1/2022.emnlp-main.155

null

27,285

inproceedings

yao-etal-2022-d4

D4: a {C}hinese Dialogue Dataset for Depression-Diagnosis-Oriented Chat

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.emnlp-main.156/

Yao, Binwei and Shi, Chao and Zou, Likai and Dai, Lingfeng and Wu, Mengyue and Chen, Lu and Wang, Zhen and Yu, Kai

Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

2438--2459

In a depression-diagnosis-directed clinical session, doctors initiate a conversation with ample emotional support that guides the patients to expose their symptoms based on clinical diagnosis criteria. Such a dialogue system is distinguished from existing single-purpose human-machine dialog systems, as it combines task-oriented and chit-chats with uniqueness in dialogue topics and procedures. However, due to the social stigma associated with mental illness, the dialogue data related to depression consultation and diagnosis are rarely disclosed. Based on clinical depression diagnostic criteria ICD-11 and DSM-5, we designed a 3-phase procedure to construct D$^4$: a Chinese Dialogue Dataset for Depression-Diagnosis-Oriented Chat, which simulates the dialogue between doctors and patients during the diagnosis of depression, including diagnosis results and symptom summary given by professional psychiatrists for each conversation. Upon the newly-constructed dataset, four tasks mirroring the depression diagnosis process are established: response generation, topic prediction, dialog summary, and severity classification of depressive episode and suicide risk. Multi-scale evaluation results demonstrate that a more empathy-driven and diagnostic-accurate consultation dialogue system trained on our dataset can be achieved compared to rule-based bots.

null

10.18653/v1/2022.emnlp-main.156

null

27,286