论文检索

Chong Li, Shaonan Wang, Yunhao Zhang, Jiajun Zhang, Chengqing Zong

Transformer-based models, even though achieving super-human performance on several downstream tasks, are often regarded as a black box and used as a whole. It is still unclear what mechanisms they have learned, especially their core module: multi-head attention. Inspired by functional specialization in the human brain, which helps to efficiently handle multiple tasks, this work attempts to figure out whether the multi-head attention module will evolve similar function separation under multi-tasking training. If it is, can this mechanism further improve the model performance? To investigate these questions, we introduce an interpreting method to quantify the degree of functional specialization in multi-head attention. We further propose a simple multi-task training method to increase functional specialization and mitigate negative information transfer in multi-task learning. Experimental results on seven pre-trained transformer models have demonstrated that multi-head attention does evolve functional specialization phenomenon after multi-task training which is affected by the similarity of tasks. Moreover, the multi-task training strategy based on functional specialization boosts performance in both multi-task learning and transfer learning without adding any parameters.

Anthony Liu, Lajanugen Logeswaran, Sungryull Sohn, Honglak Lee

Planning is an important capability of artificial agents that perform long-horizon tasks in real-world environments. In this work, we explore the use of pre-trained language models (PLMs) to reason about plan sequences from text instructions in embodied visual environments. Prior PLM based approaches for planning either assume observations are available in the form of text by a captioning model, reason about plans from the instruction alone, or incorporate information about the visual environment in limited ways (such as a pre-trained affordance function). In contrast, we show that the PLM can accurately plan even when observations are directly encoded as input prompts for the PLM. We show this simple approach outperforms prior approaches in experiments on the ALFWorld and VirtualHome benchmarks.

Siyuan Wang, Zheng Liu, Bo Peng

Medical report generation, focusing on automatically generating accurate clinical findings from medical images, is an important medical artificial intelligence task. It reduces the workload of physicians in writing reports. Many of the current methods depend heavily on labeled datasets that include a large amount of image-report pairs, but such datasets labeled by physicians are hard to acquire in clinical practice. To this end, in this paper, we introduce a self-training framework named REMOTE (i.e., Revisiting sElf-training for Medical repOrT gEneration) to exploit the unlabeled medical images and a reference-free evaluation metric MedCLIPScore to augment a small-scale medical report generation dataset for training accurate medical report generation model. Experiments and analysis conducted on the MIMIC-CXR and IU-Xray benchmark datasets demonstrate that, our REMOTE framework, using 1% labeled training data, achieves competitive performance with previous fully-supervised models that are trained on entire training data.

Shahar Jacob, Chen Shani, Dafna Shahaf

Analogy is one of the core capacities of human cognition; when faced with new situations, we often transfer prior experience from other domains. Most work on computational analogy relies heavily on complex, manually crafted input. In this work, we relax the input requirements, requiring only names of entities to be mapped. We automatically extract commonsense representations and use them to identify a mapping between the entities. Unlike previous works, our framework can handle partial analogies and suggest new entities to be added. Moreover, our method’s output is easily interpretable, allowing for users to understand why a specific mapping was chosen. Experiments show that our model correctly maps 81.2% of classical 2x2 analogy problems (guess level=50%). On larger problems, it achieves 77.8% accuracy (mean guess level=13.1%). In another experiment, we show our algorithm outperforms human performance, and the automatic suggestions of new entities resemble those suggested by humans. We hope this work will advance computational analogy by paving the way to more flexible, realistic input requirements, with broader applicability.

Molly Petersen, Lonneke van der Plas

While analogies are a common way to evaluate word embeddings in NLP, it is also of interest to investigate whether or not analogical reasoning is a task in itself that can be learned. In this paper, we test several ways to learn basic analogical reasoning, specifically focusing on analogies that are more typical of what is used to evaluate analogical reasoning in humans than those in commonly used NLP benchmarks. Our experiments find that models are able to learn analogical reasoning, even with a small amount of data. We additionally compare our models to a dataset with a human baseline, and find that after training models approach human performance.

Dian Yu, Xiaoyang Wang, Wanshun Chen, Nan Du, Longyue Wang, Haitao Mi, Dong Yu

Nonverbal messages (NM) such as speakers’ facial expressions and speed of speech are essential for face-to-face communication, and they can be regarded as implicit knowledge as they are usually not included in existing dialogue understanding or generation tasks. This paper introduces the task of extracting NMs in written text and generating NMs for spoken text. Previous studies merely focus on extracting NMs from relatively small-scale well-structured corpora such as movie scripts wherein NMs are enclosed in parentheses by scriptwriters, which greatly decreases the difficulty of extraction. To enable extracting NMs from unstructured corpora, we annotate the first NM extraction dataset for Chinese based on novels and develop three baselines to extract single-span or multi-span NM of a target utterance from its surrounding context. Furthermore, we use the extractors to extract 749K (context, utterance, NM) triples from Chinese novels and investigate whether we can use them to improve NM generation via semi-supervised learning. Experimental results demonstrate that the automatically extracted triples can serve as high-quality augmentation data of clean triples extracted from scripts to generate more relevant, fluent, valid, and factually consistent NMs than the purely supervised generator, and the resulting generator can in turn help Chinese dialogue understanding tasks such as dialogue machine reading comprehension and emotion classification by simply adding the predicted “unspoken” NM to each utterance or narrative in inputs.

Minsik Oh, Joosung Lee, Jiwei Li, Guoyin Wang

Identifying relevant persona or knowledge for conversational systems is critical to grounded dialogue response generation. However, each grounding has been mostly researched in isolation with more practical multi-context dialogue tasks introduced in recent works. We define Persona and Knowledge Dual Context Identification as the task to identify persona and knowledge jointly for a given dialogue, which could be of elevated importance in complex multi-context dialogue settings. We develop a novel grounding retrieval method that utilizes all contexts of dialogue simultaneously. Our method requires less computational power via utilizing neural QA retrieval models. We further introduce our novel null-positive rank test which measures ranking performance on semantically dissimilar samples (i.e. hard negatives) in relation to data augmentation.

Steven Wang, Antoine Scardigli, Leonard Tang, Wei Chen, Dmitry Levkin, Anya Chen, Spencer Ball, Thomas Woodside, Oliver Zhang, Dan Hendrycks

Reading comprehension of legal text can be a particularly challenging task due to the length and complexity of legal clauses and a shortage of expert-annotated datasets. To address this challenge, we introduce the Merger Agreement Understanding Dataset (MAUD), an expert-annotated reading comprehension dataset based on the American Bar Association’s 2021 Public Target Deal Points Study, with over 39,000 examples and over 47,000 total annotations. Our fine-tuned Transformer baselines show promising results, with models performing well above random on most questions. However, on a large subset of questions, there is still room for significant improvement. As the only expert-annotated merger agreement dataset, MAUD is valuable as a benchmark for both the legal profession and the NLP community.

Yixin Liu, Alexander Fabbri, Yilun Zhao, Pengfei Liu, Shafiq Joty, Chien-Sheng Wu, Caiming Xiong, Dragomir Radev

Interpretability and efficiency are two important considerations for the adoption of neural automatic metrics. In this work, we develop strong-performing automatic metrics for reference-based summarization evaluation, based on a two-stage evaluation pipeline that first extracts basic information units from one text sequence and then checks the extracted units in another sequence. The metrics we developed include two-stage metrics that can provide high interpretability at both the fine-grained unit level and summary level, and one-stage metrics that achieve a balance between efficiency and interpretability. We make the developed tools publicly available at https://github.com/Yale-LILY/AutoACU.

Mingyang Song, Pengyu Xu, Yi Feng, Huafeng Liu, Liping Jing

Over-generation errors occur when a keyphrase extraction model correctly determines a candidate keyphrase as a keyphrase because it contains a word that frequently appears in the document but at the same time erroneously outputs other candidates as keyphrases because they contain the same word. To mitigate this issue, we propose a new heterogeneous centrality detection approach (CentralityRank), which extracts keyphrases by simultaneously identifying both implicit and explicit centrality within a heterogeneous graph as the importance score of each candidate. More specifically, CentralityRank detects centrality by taking full advantage of the content within the input document to construct graphs that encompass semantic nodes of varying granularity levels, not limited to just phrases. These additional nodes act as intermediaries between candidate keyphrases, enhancing cross-phrase relations. Furthermore, we introduce a novel adaptive boundary-aware regularization that can leverage the position information of candidate keyphrases, thus influencing the importance of candidate keyphrases. Extensive experimental results demonstrate the superiority of CentralityRank over recent state-of-the-art unsupervised keyphrase extraction baselines across three benchmark datasets.

Sarah Alsayyahi, Riza Batista-Navarro

Temporal relation extraction models have thus far been hindered by a number of issues in existing temporal relation-annotated news datasets, including: (1) low inter-annotator agreement due to the lack of specificity of their annotation guidelines in terms of what counts as a temporal relation; (2) the exclusion of long-distance relations within a given document (those spanning across different paragraphs); and (3) the exclusion of events that are not centred on verbs. This paper aims to alleviate these issues by presenting a new annotation scheme that clearly defines the criteria based on which temporal relations should be annotated. Additionally, the scheme includes events even if they are not expressed as verbs (e.g., nominalised events). Furthermore, we propose a method for annotating all temporal relations—including long-distance ones—which automates the process, hence reducing time and manual effort on the part of annotators. The result is a new dataset, the TIMELINE corpus, in which improved inter-annotator agreement was obtained, in comparison with previously reported temporal relation datasets. We report the results of training and evaluating two baseline temporal relation extraction models on the new corpus, and compare them with results obtained on the widely used MATRES corpus.

Iulia Comsa, Srini Narayanan

Spatial reasoning is a fundamental building block of human cognition, used in representing, grounding, and reasoning about physical and abstract concepts. We propose a novel benchmark focused on assessing inferential properties of statements with spatial prepositions. The benchmark includes original datasets in English and Romanian and aims to probe the limits of reasoning about spatial relations in large language models. We use prompt engineering to study the performance of two families of large language models, PaLM and GPT-3, on our benchmark. Our results show considerable variability in the performance of smaller and larger models, as well as across prompts and languages. However, none of the models reaches human performance.

Jb. Kim, Hazel Kim, Joonghyuk Hahn, Yo-Sub Han

Solving math word problems depends on how to articulate the problems, the lens through which models view human linguistic expressions. Real-world settings count on such a method even more due to the diverse practices of the same mathematical operations. Earlier works constrain available thinking processes by limited prediction strategies without considering their significance in acquiring mathematical knowledge. We introduce Attention-based THought Expansion Network Architecture (ATHENA) to tackle the challenges of real-world practices by mimicking human thought expansion mechanisms in the form of neural network propagation. A thought expansion recurrently generates the candidates carrying the thoughts of possible math expressions driven from the previous step and yields reasonable thoughts by selecting the valid pathways to the goal. Our experiments show that ATHENA achieves a new state-of-the-art stage toward the ideal model that is compelling in variant questions even when the informativeness in training examples is restricted.

Shushan Arakelyan, Rocktim Das, Yi Mao, Xiang Ren

We systematically study how three large language models with code capabilities - CodeT5, Codex, and ChatGPT - generalize to out-of-domain data. We consider two fundamental applications - code summarization, and code generation. We split data into domains following its natural boundaries - by an organization, by a project, and by a module within the software project. We establish that samples from each new domain present all the models with a significant challenge of distribution shift. We study how established methods adapt models to better generalize to new domains. Our experiments show that while multitask learning alone is a reasonable baseline, combining it with few-shot finetuning on examples retrieved from training data can achieve very strong performance. Moreover, this solution can outperform direct finetuning for very low-data scenarios. Finally, we consider variations of this approach to create a more broadly applicable method to adapt to multiple domains at once. We find that for code generation, a model adapted to multiple domains simultaneously performs on par with those adapted to a single domain.

Michael Wiegand, Jana Kampfmeier, Elisabeth Eder, Josef Ruppenhofer

We address the task of identifying euphemistic abuse (e.g. “You inspire me to fall asleep”) paraphrasing simple explicitly abusive utterances (e.g. “You are boring”). For this task, we introduce a novel dataset that has been created via crowdsourcing. Special attention has been paid to the generation of appropriate negative (non-abusive) data. We report on classification experiments showing that classifiers trained on previous datasets are less capable of detecting such abuse. Best automatic results are obtained by a classifier that augments training data from our new dataset with automatically-generated GPT-3 completions. We also present a classifier that combines a few manually extracted features that exemplify the major linguistic phenomena constituting euphemistic abuse.

Dominic Petrak, Nafise Moosavi, Ye Tian, Nikolai Rozanov, Iryna Gurevych

Learning from free-text human feedback is essential for dialog systems, but annotated data is scarce and usually covers only a small fraction of error types known in conversational AI. Instead of collecting and annotating new datasets from scratch, recent advances in synthetic dialog generation could be used to augment existing dialog datasets with the necessary annotations. However, to assess the feasibility of such an effort, it is important to know the types and frequency of free-text human feedback included in these datasets. In this work, we investigate this question for a variety of commonly used dialog datasets, including MultiWoZ, SGD, BABI, PersonaChat, Wizardsof-Wikipedia, and the human-bot split of the Self-Feeding Chatbot. Using our observations, we derive new taxonomies for the annotation of free-text human feedback in dialogs and investigate the impact of including such data in response generation for three SOTA language generation models, including GPT-2, LLAMA, and Flan-T5. Our findings provide new insights into the composition of the datasets examined, including error types, user response types, and the relations between them.

Yoo Yeon Sung, Jordan Boyd-Graber, Naeemul Hassan

Polarization and the marketplace for impressions have conspired to make navigating information online difficult for users, and while there has been a significant effort to detect false or misleading text, multimodal datasets have received considerably less attention. To complement existing resources, we present multimodal Video Misleading Headline (VMH), a dataset that consists of videos and whether annotators believe the headline is representative of the video’s contents. After collecting and annotating this dataset, we analyze multimodal baselines for detecting misleading headlines. Our annotation process also focuses on why annotators view a video as misleading, allowing us to better understand the interplay of annotators’ background and the content of the videos.

Ankita Bhaumik, Praveen Venkateswaran, Yara Rizk, Vatche Isahagian

The popularity of conversational digital assistants has resulted in the availability of large amounts of conversational data which can be utilized for improved user experience and personalized response generation. Building these assistants using popular large language models like ChatGPT also require additional emphasis on prompt engineering and evaluation methods. Textual similarity metrics are a key ingredient for such analysis and evaluations. While many similarity metrics have been proposed in the literature, they have not proven effective for task-oriented conversations as they do not take advantage of unique conversational features. To address this gap, we present TaskDiff, a novel conversational similarity metric that utilizes different dialogue components (utterances, intents, and slots) and their distributions to compute similarity. Extensive experimental evaluation of TaskDiff on a benchmark dataset demonstrates its superior performance and improved robustness over other related approaches.

Souvika Sarkar, Dongji Feng, Shubhra Kanti Karmaker Santu

In this paper, we conducted a comprehensive study with the latest Sentence Encoders and Large Language Models (LLMs) on the challenging task of “definition-wild zero-shot topic inference”, where users define or provide the topics of interest in real-time. Through extensive experimentation on seven diverse data sets, we observed that LLMs, such as ChatGPT-3.5 and PaLM, demonstrated superior generality compared to other LLMs, e.g., BLOOM and GPT-NeoX. Furthermore, Sentence-BERT, a BERT-based classical sentence encoder, outperformed PaLM and achieved performance comparable to ChatGPT-3.5.

Chi Seng Cheang, Hou Pong Chan, Derek F. Wong, Xuebo Liu, Zhaocong Li, Yanming Sun, Shudong Liu, Lidia S. Chao

Recent pre-trained language models (PLMs) achieve promising results in existing abstractive summarization datasets. However, existing summarization benchmarks overlap in time with the standard pre-training corpora and finetuning datasets. Hence, the strong performance of PLMs may rely on the parametric knowledge that is memorized during pre-training and fine-tuning. Moreover, the knowledge memorized by PLMs may quickly become outdated, which affects the generalization performance of PLMs on future data. In this work, we propose TempoSum, a novel benchmark that contains data samples from 2010 to 2022, to understand the temporal generalization ability of abstractive summarization models. Through extensive human evaluation, we show that parametric knowledge stored in summarization models significantly affects the faithfulness of the generated summaries on future data. Moreover, existing faithfulness enhancement methods cannot reliably improve the faithfulness of summarization models on future data. Finally, we discuss several recommendations to the research community on how to evaluate and improve the temporal generalization capability of text summarization models.