EMNLP 2022 Papers

Prominent questions about the role of sensory vs. linguistic input in the way we acquire and use language have been extensively studied in the psycholinguistic literature. However, the relative effect of various factors in a person’s overall experience on their linguistic system remains unclear. We study this question by making a step forward towards a better understanding of the conceptual perception of colors by color-blind individuals, as reflected in their spontaneous linguistic productions. Using a novel and carefully curated dataset, we show that red-green color-blind speakers use the “red” and “green” color terms in less predictable contexts, and in linguistic environments evoking mental image to a lower extent, when compared to their normal-sighted counterparts. These findings shed some new and interesting light on the role of sensory experience on our linguistic system.

Collecting dialogue data with domain-slot-value labels for dialogue state tracking (DST) could be a costly process. In this paper, we propose a novel framework based on domain-slot related description to tackle the challenge of few-shot cross-domain DST. Specifically, we design an extraction module to extract domain-slot related verbs and nouns in the dialogue. Then, we integrates them into the description, which aims to prompt the model to identify the slot information. Furthermore, we introduce a random sampling strategy to improve the domain generalization ability of the model. We utilize a pre-trained model to encode contexts and description and generates answers with an auto-regressive manner. Experimental results show that our approaches substantially outperform the existing few-shot DST methods on MultiWOZ and gain strong improvements on the slot accuracy comparing to existing slot description methods.

Automatic Peer-review Aspect Score Prediction (PASP) of academic papers can be a helpful assistant tool for both reviewers and authors. Most existing works on PASP utilize supervised learning techniques. However, the limited number of peer-review data deteriorates the performance of PASP. This paper presents a novel semi-supervised learning (SSL) method that incorporates the Transformer fine-tuning into the Γ-model, a variant of the Ladder network, to leverage contextual features from unlabeled data. Backpropagation simultaneously minimizes the sum of supervised and unsupervised cost functions, avoiding the need for layer-wise pre-training. The experimental results show that our model outperforms the supervised and naive semi-supervised learning baselines. Our source codes are available online.

Code-switching (CS) is a very common phenomenon in regions with various co-existing languages. Since CS is such a frequent habit in informal communications, both spoken and written, it also arises naturally in Human-Machine Interactions. Therefore, in order for natural language understanding (NLU) not to be degraded, CS must be taken into account when developing chatbots. The co-existence of multiple languages in a single NLU model has become feasible with multilingual language representation models such as mBERT. In this paper, the efficacy of zero-shot cross-lingual transfer learning with mBERT for NLU is evaluated on a Basque-Spanish CS chatbot corpus, comparing the performance of NLU models trained using in-domain chatbot utterances in Basque and/or Spanish without CS. The results obtained indicate that training joint multi-intent classification and entity recognition models on both languages simultaneously achieves best performance, better capturing the CS patterns.

Hierarchical text classification aims to leverage label hierarchy in multi-label text classification. Existing methods encode label hierarchy in a global view, where label hierarchy is treated as the static hierarchical structure containing all labels. Since global hierarchy is static and irrelevant to text samples, it makes these methods hard to exploit hierarchical information. Contrary to global hierarchy, local hierarchy as a structured labels hierarchy corresponding to each text sample. It is dynamic and relevant to text samples, which is ignored in previous methods. To exploit global and local hierarchies, we propose Hierarchy-guided BERT with Global and Local hierarchies (HBGL), which utilizes the large-scale parameters and prior language knowledge of BERT to model both global and local hierarchies. Moreover, HBGL avoids the intentional fusion of semantic and hierarchical modules by directly modeling semantic and hierarchical information with BERT. Compared with the state-of-the-art method HGCLR, our method achieves significant improvement on three benchmark datasets.

Event argument extraction is a challenging subtask of event extraction, aiming to identify and assign roles to arguments under a certain event. Existing methods extract arguments of each role independently, ignoring the relationship between different roles. Such an approach hinders the model from learning explicit interactions between different roles to improve the performance of individual argument extraction. As a solution, we design a neural model that we refer to as the Explicit Role Interaction Network (ERIN) which allows for dynamically capturing the correlations between different argument roles within an event. Extensive experiments on the benchmark dataset ACE2005 demonstrate the superiority of our proposed model to existing approaches.

In a conversational search scenario, a query might be context-dependent because some words are referred to previous expressions or omitted. Previous works tackle the issue by either reformulating the query into a self-contained query (query rewriting) or learning a contextualized query embedding from the query context (context modelling). In this paper, we propose a model CRDR that can perform query rewriting and context modelling in a unified framework in which the query rewriting’s supervision signals further enhance the context modelling. Instead of generating a new query, CRDR only performs necessary modifications on the original query, which improves both accuracy and efficiency of query rewriting. In the meantime, the query rewriting benefits the context modelling by explicitly highlighting relevant terms in the query context, which improves the quality of the learned contextualized query embedding. To verify the effectiveness of CRDR, we perform comprehensive experiments on TREC CAsT-19 and TREC CAsT-20 datasets, and the results show that our method outperforms all baseline models in terms of both quality of query rewriting and quality of context-aware ranking.

Joint intent detection and slot filling is a key research topic in natural language understanding (NLU). Existing joint intent and slot filling systems analyze and compute features collectively for all slot types, and importantly, have no way to explain the slot filling model decisions. In this work, we propose a novel approach that: (i) learns to generate additional slot type specific features in order to improve accuracy and (ii) provides explanations for slot filling decisions for the first time in a joint NLU model. We perform an additional constrained supervision using a set of binary classifiers for the slot type specific feature learning, thus ensuring appropriate attention weights are learned in the process to explain slot filling decisions for utterances. Our model is inherently explainable and does not need any post-hoc processing. We evaluate our approach on two widely used datasets and show accuracy improvements. Moreover, a detailed analysis is also provided for the exclusive slot explainability.

Multi-hop Question Answering is an agent task for testing the reasoning ability. With the development of pre-trained models, the implicit reasoning ability has been surprisingly improved and can even surpass human performance. However, the nature of the black box hinders the construction of explainable intelligent systems. Several researchers have explored explainable neural-symbolic reasoning methods based on question decomposition techniques. The undifferentiable symbolic operations and the error propagation in the reasoning process lead to poor performance. To alleviate it, we propose a simple yet effective Global Differentiable Learning strategy to explore optimal reasoning paths from the latent probability space so that the model learns to solve intermediate reasoning processes without expert annotations. We further design a Dynamic Adaptive Reasoner to enhance the generalization of unseen questions. Our method achieves 17% improvements in F1-score against BreakRC and shows better interpretability. We take a step forward in building interpretable reasoning methods.

Expert Finding is an important task in Community Question Answering (CQA) platforms, which could help route questions to potential users to answer. The key is to learn representations of experts based on their historical answered questions accurately. In this paper, inspired by the strong text understanding ability of Pretrained Language modelings (PLMs), we propose a pre-training and fine-tuning expert finding framework. The core is that we design an expert-level pre-training paradigm, that effectively integrates expert interest and expertise simultaneously. Specifically different from the typical corpus-level pre-training, we treat each expert as the basic pre-training unit including all the historical answered question titles of the expert, which could fully indicate the expert interests for questions. Besides, we integrate the vote score information along with each answer of the expert into the pre-training phrase to model the expert ability explicitly. Finally, we propose a novel reputation-augmented Masked Language Model (MLM) pre-training strategy to capture the expert reputation information. In this way, our method could learn expert representation comprehensively, which then will be adopted and fine-tuned in the down-streaming expert-finding task. Extensive experimental results on six real-world CQA datasets demonstrate the effectiveness of our method.

The field of Deep Learning (DL) has undergone explosive growth during the last decade, with a substantial impact on Natural Language Processing (NLP) as well. Yet, compared to more established disciplines, a lack of common experimental standards remains an open challenge to the field at large. Starting from fundamental scientific principles, we distill ongoing discussions on experimental standards in NLP into a single, widely-applicable methodology. Following these best practices is crucial to strengthen experimental evidence, improve reproducibility and enable scientific progress. These standards are further collected in a public repository to help them transparently adapt to future needs.

The tasks of humor understanding and generation are challenging and subjective even for humans, requiring commonsense and real-world knowledge to master. Puns, in particular, add the challenge of fusing that knowledge with the ability to interpret lexical-semantic ambiguity. In this paper, we present the ExPUNations (ExPUN) dataset, in which we augment an existing dataset of puns with detailed crowdsourced annotations of keywords denoting the most distinctive words that make the text funny, pun explanations describing why the text is funny, and fine-grained funniness ratings. This is the first humor dataset with such extensive and fine-grained annotations specifically for puns. Based on these annotations, we propose two tasks: explanation generation to aid with pun classification and keyword-conditioned pun generation, to challenge the current state-of-the-art natural language understanding and generation models’ ability to understand and generate humor. We showcase that the annotated keywords we collect are helpful for generating better novel humorous texts in human evaluation, and that our natural language explanations can be leveraged to improve both the accuracy and robustness of humor classifiers.

With the increase in availability of large pre-trained language models (LMs) in Natural Language Processing (NLP), it becomes critical to assess their fit for a specific target task a priori—as fine-tuning the entire space of available LMs is computationally prohibitive and unsustainable. However, encoder transferability estimation has received little to no attention in NLP. In this paper, we propose to generate quantitative evidence to predict which LM, out of a pool of models, will perform best on a target task without having to fine-tune all candidates. We provide a comprehensive study on LM ranking for 10 NLP tasks spanning the two fundamental problem types of classification and structured prediction. We adopt the state-of-the-art Logarithm of Maximum Evidence (LogME) measure from Computer Vision (CV) and find that it positively correlates with final LM performance in 94% of the setups.In the first study of its kind, we further compare transferability measures with the de facto standard of human practitioner ranking, finding that evidence from quantitative metrics is more robust than pure intuition and can help identify unexpected LM candidates.

Human reading comprehension often requires reasoning of event semantic relations in narratives, represented by Event-centric Question-Answering (QA). To address event-centric QA, we propose a novel QA model with contrastive learning and invertible event transformation, call TranCLR. Our proposed model utilizes an invertible transformation matrix to project semantic vectors of events into a common event embedding space, trained with contrastive learning, and thus naturally inject event semantic knowledge into mainstream QA pipelines. The transformation matrix is fine-tuned with the annotated event relation types between events that occurred in questions and those in answers, using event-aware question vectors. Experimental results on the Event Semantic Relation Reasoning (ESTER) dataset show significant improvements in both generative and extractive settings compared to the existing strong baselines, achieving over 8.4% gain in the token-level F1 score and 3.0% gain in Exact Match (EM) score under the multi-answer setting. Qualitative analysis reveals the high quality of the generated answers by TranCLR, demonstrating the feasibility of injecting event knowledge into QA model learning. Our code and models can be found at https://github.com/LuJunru/TranCLR.

The automatic generation of Multiple Choice Questions (MCQ) has the potential to reduce the time educators spend on student assessment significantly. However, existing evaluation metrics for MCQ generation, such as BLEU, ROUGE, and METEOR, focus on the n-gram based similarity of the generated MCQ to the gold sample in the dataset and disregard their educational value.They fail to evaluate the MCQ’s ability to assess the student’s knowledge of the corresponding target fact. To tackle this issue, we propose a novel automatic evaluation metric, coined Knowledge Dependent Answerability (KDA), which measures the MCQ’s answerability given knowledge of the target fact. Specifically, we first show how to measure KDA based on student responses from a human survey.Then, we propose two automatic evaluation metrics, KDA_disc and KDA_cont, that approximate KDA by leveraging pre-trained language models to imitate students’ problem-solving behavior.Through our human studies, we show that KDA_disc and KDA_soft have strong correlations with both (1) KDA and (2) usability in an actual classroom setting, labeled by experts. Furthermore, when combined with n-gram based similarity metrics, KDA_disc and KDA_cont are shown to have a strong predictive power for various expert-labeled MCQ quality measures.

Despite their strong performance on many tasks, pre-trained language models have been shown to struggle on out-of-distribution compositional generalization. Meanwhile, recent work has shown considerable improvements on many NLP tasks from model scaling. Can scaling up model size also improve compositional generalization in semantic parsing? We evaluate encoder-decoder models up to 11B parameters and decoder-only models up to 540B parameters, and compare model scaling curves for three different methods for applying a pre-trained language model to a new task: fine-tuning all parameters, prompt tuning, and in-context learning. We observe that fine-tuning generally has flat or negative scaling curves on out-of-distribution compositional generalization in semantic parsing evaluations. In-context learning has positive scaling curves, but is generally outperformed by much smaller fine-tuned models. Prompt-tuning can outperform fine-tuning, suggesting further potential improvements from scaling as it exhibits a more positive scaling curve. Additionally, we identify several error trends that vary with model scale. For example, larger models are generally better at modeling the syntax of the output space, but are also more prone to certain types of overfitting. Overall, our study highlights limitations of current techniques for effectively leveraging model scale for compositional generalization, while our analysis also suggests promising directions for future work.

To explain NLP models a popular approach is to use importance measures, such as attention, which inform input tokens are important for making a prediction. However, an open question is how well these explanations accurately reflect a model’s logic, a property called faithfulness. To answer this question, we propose Recursive ROAR, a new faithfulness metric. This works by recursively masking allegedly important tokens and then retraining the model. The principle is that this should result in worse model performance compared to masking random tokens. The result is a performance curve given a masking-ratio. Furthermore, we propose a summarizing metric using area-between-curves (ABC), which allows for easy comparison across papers, models, and tasks. We evaluate 4 different importance measures on 8 different datasets, using both LSTM-attention models and RoBERTa models. We find that the faithfulness of importance measures is both model-dependent and task-dependent. This conclusion contradicts previous evaluations in both computer vision and faithfulness of attention literature.

Current metrics for evaluating factuality for abstractive document summarization have achieved high correlations with human judgment, but they do not account for the vision modality and thus are not adequate for vision-and-language summarization. We propose CLIPBERTSCORE, a simple weighted combination of CLIPScore and BERTScore to leverage the robustness and strong factuality detection performance between image-summary and document-summary, respectively. Next, due to the lack of meta-evaluation benchmarks to evaluate the quality of multimodal factuality metrics, we collect human judgments of factuality with respect to documents and images. We show that this simple combination of two metrics in the zero-shot setting achieves higher correlations than existing factuality metrics for document summarization, outperforms an existing multimodal summarization metric, and performs competitively with strong multimodal factuality metrics specifically fine-tuned for the task. Our thorough analysis demonstrates the robustness and high correlation of CLIPBERTSCORE and its components on four factuality metric-evaluation benchmarks. Finally, we demonstrate two practical downstream applications of our CLIPBERTSCORE metric: for selecting important images to focus on during training, and as a reward for reinforcement learning to improve factuality of multimodal summary generation w.r.t automatic and human evaluation.

With the recent success of dense retrieval methods based on bi-encoders, studies have applied this approach to various interesting downstream retrieval tasks with good efficiency and in-domain effectiveness.Recently, we have also seen the presence of dense retrieval models in Math Information Retrieval (MIR) tasks,but the most effective systems remain classic retrieval methods that consider hand-crafted structure features.In this work, we try to combine the best of both worlds: a well-defined structure search method for effective formula search and efficient bi-encoder dense retrieval models to capture contextual similarities.Specifically, we have evaluated two representative bi-encoder models for token-level and passage-level dense retrieval on recent MIR tasks.Our results show that bi-encoder models are highly complementary to existing structure search methods, and we are able to advance the state-of-the-art on MIR datasets.