Papers by Yoshihiko Hayashi
Towards Answer-unaware Conversational Question Generation (D19-58)
Copied to clipboard
| Challenge: | Existing frameworks for conversational question generation are answeraware, but are not able to generate corresponding answers . a number of question generation methods are developed for text-based question answering . |
| Approach: | They propose a framework for conversational question generation that is unaware of the corresponding answers. |
| Outcome: | The proposed framework is effective but answeraware, the authors show . it improves quality of generated questions if question foci and question patterns are identified . |
Reassessing Semantic Knowledge Encoded in Large Language Models through the Word-in-Context Task (2024.lrec-main)
Copied to clipboard
| Challenge: | Recent advances in large language models (LLMs) have propelled significant progress, extending their application across various domains including dialogue systems, text generation, translation systems, and beyond. |
| Approach: | They propose to use the Word-in-Context (WiC) task to reassess the semantic knowledge encoded in large language models (LLMs) they prompt LLMs to generate natural language descriptions that contrast the meanings of the target word in two contextual sentences given in the WiC dataset. |
| Outcome: | The proposed model significantly improves the classification accuracy of the two models. |
Answerable or Not: Devising a Dataset for Extending Machine Reading Comprehension (C18-1)
Copied to clipboard
| Challenge: | Existing MRC algorithms assume that each question is answerable by looking at text passages, but to realize human-like language comprehension ability, a machine should be able to distinguish not-answerable questions from answerable questions. |
| Approach: | They propose a method for automatically assigning difficulty level labels to a dataset that alters an existing MRC dataset and describes the resulting dataset. |
| Outcome: | The proposed method can detect NAQs in a dataset with difficulty level labels and is valid and potentially useful in the development of advanced MRC models. |
Phrase-Level Localization of Inconsistency Errors in Summarization by Weak Supervision (2022.coling-1)
Copied to clipboard
| Challenge: | Existing methods for evaluating inconsistency in summarization are limited . a recent study found that more than 30% of summarized summaries are inconsistent with the source documents . |
| Approach: | They propose a method for localizing inconsistency errors in summarization using a synthetic dataset that contains factual errors likely to be produced by a common language processor. |
| Outcome: | The proposed method detects factual errors more accurately than existing weakly supervised methods . the proposed model also detects errors in original sentences more accurately . |
Social Image Tags as a Source of Word Embeddings: A Task-oriented Evaluation (L18-1)
Copied to clipboard
| Challenge: | Distributional hypothesis-based word representations lack perceptual and empirical knowledge. |
| Approach: | They evaluate the effectiveness of social image tags in generating word embeddings . they find that generated word embeds exhibit somewhat different behaviors from corpus-originated representations - authors . |
| Outcome: | The generated word embeddings exhibit comparable performance with corpus-originated representations. |
Exploiting Narrative Context and A Priori Knowledge of Categories in Textual Emotion Classification (2020.coling-main)
Copied to clipboard
| Challenge: | Existing methods for recognizing the mental state of characters in text are limited by their use of character-specific contexts. |
| Approach: | They propose a method that encodes the preceding context of the target sentence along with the target phrase using a BERT-based text encoder. |
| Outcome: | The proposed method improves the accuracy of emotion classification by encoding the preceding context of the target sentence along with the target phrase using a BERT encoder. |
Evaluating the Effects of Embedding with Speaker Identity Information in Dialogue Summarization (2022.lrec-1)
Copied to clipboard
| Challenge: | Existing methods for automatic dialogue summarization do not take into account speaker identity information, but instead use sinusoidal functions to embed speaker information at the less informative part of the position embedding. |
| Approach: | They propose to embed speaker identity information into a dialogue transcript encoder to address this issue and reduce the "who said what"-related errors. |
| Outcome: | The proposed method improves the convergence of the model in training and increases the average ROUGE scores of the generated summaries in comparison to existing methods. |
Word Attribute Prediction Enhanced by Lexical Entailment Tasks (2020.lrec-1)
Copied to clipboard
| Challenge: | a semantic attribute is associated with a designated dimension in attribute-based vector representations . semantic attributes are created by psychological experimental settings involving human annotators . a conceptual attribute of a concept dictates a specific semantic aspect of the concept . |
| Approach: | They propose a two-stage neural network architecture that fine-tunes attribute representations by employing supervised entailment tasks. |
| Outcome: | The proposed method improves performance of semantic/visual similarity/relatedness evaluation tasks. |
Towards the Detection of a Semantic Gap in the Chain of Commonsense Knowledge Triples (2022.lrec-1)
Copied to clipboard
| Challenge: | a commonsense knowledge resource organizes common sense that is not necessarily correct all the time, but most people are expected to know or believe. |
| Approach: | They propose a machine learning-based approach to detect semantic gaps in a commonsense knowledge graph . they use a conceptNet dataset to test the validity of two adjacent triples . |
| Outcome: | The proposed approach detects a semantic gap in a commonsense knowledge graph . the proposed approach also provides insights into the effectiveness of sense embeddings . |
Evaluating LLMs’ Capability to Identify Lexical Semantic Equivalence: Probing with the Word-in-Context Task (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods to evaluate the capability of large language models to identify lexical semantic equivalence are not currently being used. |
| Approach: | They propose to use the Word-in-Context (WiC) task to determine whether the meanings of a target word remain identical across different contexts to evaluate their capability. |
| Outcome: | The proposed method outperforms other LLMs in the Word-in-Context (WiC) task. |