Papers by Hidetaka Kamigaito

Tomoyuki Jinno, Kazuki Hayashi, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

Challenge:	Existing studies show that MBR decoding improves model generation performance . however, the theoretical underpinnings of these results remain uncertain .
Approach:	They propose a theoretical interpretation of MBR decoding from the perspective of bias–diversity decomposition.
Outcome:	The proposed method improves the quality estimation of hypotheses by decomposing bias and diversity into two main factors.

Copied to clipboard

Length Representations in Large Language Models (2025.findings-emnlp)

Challenge:	Existing knowledge probes for pre-trained language models exhibit quadratic time complexity, limiting the size of knowledge graphs used for probing.
Approach:	They propose an embedding-based relational probe that evaluates pre-trained language models' factual knowledge retrieval capabilities.
Outcome:	The proposed probe achieves effective time complexity of linear order O(n), supports rank-based evaluation metrics including Hit@k, handles multi-token entity names and enables probing whilst disambiguating homographic tail-entity names.

Copied to clipboard

Sangjun Moon, Dasom Choi, Jingun Kwon, Hidetaka Kamigaito, Manabu Okumura

Revisiting Non-Verbatim Memorization in Large Language Models: The Role of Entity Surface Forms (2026.acl-long)

Challenge:	Large language models (LLMs) have shown remarkable capabilities across various tasks, that are learned from massive amounts of text-based data.
Approach:	They propose to scale hidden units within the model to control output sequence length without losing the informativeness of the generated text.
Outcome:	The output sequence length is controlled by multiple head attention mechanisms, which can be adjusted in a disentangled manner.

Copied to clipboard

Yuto Nishida, Naoki Shikoda, Yosuke Kishinami, Ryo Fujii, Makoto Morishita, Hidetaka Kamigaito, Taro Watanabe

Pointing to Subwords for Generating Function Names in Source Code (2020.coling-main)

Challenge:	Entity-based QA is a common framework for analyzing non-verbatim memorization, but typically query each entity using a single canonical surface form.
Approach:	They propose a dataset that pairs Wikidata factual triples with categorized entity surface forms . they examine surface-conditioned factual memorization and find that prediction outcomes change when only the entity surface form is changed.
Outcome:	The proposed dataset shows that large language models memorize factual knowledge when only the subject entity surface form is changed.

Copied to clipboard

Shogo Fujita, Hidetaka Kamigaito, Hiroya Takamura, Manabu Okumura

A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing (2022.findings-emnlp)

Challenge:	Existing methods for generating function names from source code face difficulties in generating low-frequency or out-of-vocabulary subwords.
Approach:	They propose two strategies for copying low-frequency or out-of-vocabulary subwords in inputs.
Outcome:	The proposed method improves on the Java-small and Java-large datasets and improves the existing method on the GitHub platform.

Copied to clipboard

Naoki Kobayashi, Tsutomu Hirao, Hidetaka Kamigaito, Manabu Okumura, Masaaki Nagata

Centroid-Based Efficient Minimum Bayes Risk Decoding (2024.findings-acl)

Challenge:	Existing discourse parsing methods need a strong baseline for reporting reliable experimental results.
Approach:	They integrate existing parsing strategies with transformer-based pre-trained language models to provide a strong baseline for reporting reliable experimental results.
Outcome:	The proposed model outperforms the current best model using DeBERTa.

Copied to clipboard

Hiroyuki Deguchi, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe, Hideki Tanaka, Masao Utiyama

Tonguescape: Exploring Language Models Understanding of Vowel Articulation (2025.naacl-long)

Challenge:	Minimum Bayes risk (MBR) decoding requires quadratic time since it computes the expected score between a translation hypothesis and all reference translations.
Approach:	They propose a centroid-based MBR decoding method that clusters the translations in the feature space and calculates the expected score using the centroids of each cluster.
Outcome:	The proposed method outperforms vanilla MBR decoding in translation quality by up to 0.5 COMET in the WMT’22 EnJa, EnDe, EnZh, and WMT'23 Enja translation tasks.

Copied to clipboard

Haruki Sakajo, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

Split or Merge: Which is Better for Unsupervised RST Parsing? (D19-1)

Challenge:	a study shows that language models can explain vowel pronunciation based on tongue positions . a visual LM can explain the relationship between vowels and tongue positions, but it is unclear whether they align textual information with visual information.
Approach:	They created video and image datasets from MRI data to examine if LMs associate real tongue positions with vowel articulation.
Outcome:	The proposed model can explain vowel pronunciation and the correlation between vowels and tongue positions as textual knowledge.

Copied to clipboard

Naoki Kobayashi, Tsutomu Hirao, Kengo Nakamura, Hidetaka Kamigaito, Manabu Okumura, Masaaki Nagata

Decoding Uncertainty: The Impact of Decoding Strategies for Uncertainty Estimation in Large Language Models (2025.findings-emnlp)

Challenge:	Rhetorical Structure Theory (RST) parsers have been based on supervised learning approaches that require an annotated corpus of sufficient size and quality.
Approach:	They propose two unsupervised methods that build an optimal RST tree based on a dissimilarity score function for splitting a text span into smaller ones and a similarity score for merging two adjacent spans into a large one.
Outcome:	The proposed method achieves the best score on English and German RST treebanks, around 0.8 F1 score, close to the previous supervised parsers.

Copied to clipboard

Wataru Hashimoto, Hidetaka Kamigaito, Taro Watanabe

Aspect-based Analysis of Advertising Appeals for Search Engine Advertising (2022.naacl-industry)

Challenge:	Decoding strategies affect the probability distribution underlying the output of a language model and can therefore affect both generation quality and uncertainty.
Approach:	They investigate the impact of decoding strategies on uncertainty estimation in large language models .
Outcome:	The proposed methods improve the uncertainty estimation of large language models by reducing repetition.

Copied to clipboard

Soichiro Murakami, Peinan Zhang, Sho Hoshino, Hidetaka Kamigaito, Hiroya Takamura, Manabu Okumura

Information Retrieval

Challenge:	ad creators must consider various aspects of advertising appeals such as price, product features, and quality in their ac work.
Approach:	They propose to use a dataset of ad texts to explore the effective aspects of advertising appeals (A3) for different industries to assist a search engine ap creators.
Outcome:	The proposed model can detect aspects of ad texts and help them estimate their performance.

Are Data Augmentation Methods in Named Entity Recognition Applicable for Uncertainty Estimation? (2024.emnlp-main)

Copied to clipboard

Wataru Hashimoto, Hidetaka Kamigaito, Taro Watanabe

Unified Interpretation of Softmax Cross-Entropy and Negative Sampling: With Case Study for Knowledge Graph Embedding (2021.acl-long)

Challenge:	Named Entity Recognition (NER) is a key task in NLP to find mentions of named entities and classify them into predefined categories.
Approach:	They investigated the impact of data augmentation on confidence calibration and uncertainty estimation in Named Entity Recognition (NER) tasks.
Outcome:	The data augmentation improves calibration and uncertainty in cross-genre and cross-lingual setting, especially in-domain setting.

Copied to clipboard

Hidetaka Kamigaito, Katsuhiko Hayashi

Disentangling Pretrained Representation to Leverage Low-Resource Languages in Multilingual Machine Translation (2024.lrec-main)

Challenge:	Existing knowledge graph embedding methods do not provide a fairly accurate comparison of the two loss functions.
Approach:	They propose to use the Bregman divergence to provide a unified interpretation of the softmax cross-entropy and negative sampling loss functions.
Outcome:	The proposed model can be used to predict missing relational links between entities using a scoring method.

Copied to clipboard

Frederikus Hudi, Zhi Qu, Hidetaka Kamigaito, Taro Watanabe

Generative Replay Inspired by Hippocampal Memory Indexing for Continual Language Learning (2023.eacl-main)

Challenge:	Multilingual neural machine translation requires an enormous dataset, leaving the low-resource language (LRL) underdeveloped.
Approach:	They evaluated five languages using a parallel corpus of 1,000 instances each and found a zero-shot improvement of 7.4 from the baseline score of 7.1 to a score of 15.5 at best.
Outcome:	The proposed model improves performance in the linguistically diverse country of Indonesia by 7.4 from baseline score of 7.1 to 15.5 at best.

Copied to clipboard

Aru Maekawa, Hidetaka Kamigaito, Kotaro Funakoshi, Manabu Okumura

Information Retrieval

Challenge:	Continual learning (CL) is a fundamental requirement for human-like general intelligence (Parisi et al., 2019).
Approach:	They propose to control sample generation using compressed features of previous training samples by using hippocampal memory indexing to enhance the generative replay.
Outcome:	The proposed method outperforms current generative replay methods and generates training samples from previous tasks.

Abstractive Document Summarization with Summary-length Prediction (2023.findings-eacl)

Copied to clipboard

Jingun Kwon, Hidetaka Kamigaito, Manabu Okumura

How to Make the Most of LLMs’ Grammatical Knowledge for Acceptability Judgments (2025.naacl-long)

Challenge:	Existing abstractive summarization models do not consider summarizing-specific information such as the target summary length.
Approach:	They propose a method for enabling a model to understand summarization-specific information by predicting the summary length in the encoder and generating a summary of the predicted length in fine-tuning.
Outcome:	The proposed method improves ROUGE scores on the WikiHow, NYT, and CNN/DM datasets.

Copied to clipboard

Yusuke Ide, Yuto Nishida, Justin Vasselli, Miyu Oba, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

An Empirical Study of Building a Strong Baseline for Constituency Parsing (P18-2)

Challenge:	Conventional approaches compare sentence probabilities directly, but large language models (LLMs) provide nuanced evaluation methods using prompts and templates.
Approach:	They propose to derive acceptability judgments from large language models using prompts and templates to comprehensively evaluate their grammatical knowledge.
Outcome:	The proposed methods excel in different linguistic phenomena, suggesting they access different aspects of the LLMs’ grammatical knowledge.

Copied to clipboard

Jun Suzuki, Sho Takase, Hidetaka Kamigaito, Makoto Morishita, Masaaki Nagata

Unveiling the Power of Source: Source-based Minimum Bayes Risk Decoding for Neural Machine Translation (2025.acl-long)

Challenge:	Sequence-to-sequence models have been used for natural language generation tasks such as machine translation and summarization.
Approach:	They propose to build a strong baseline based on general purpose sequence-to-sequence models for constituency parsing.
Outcome:	The proposed model outperforms existing models in natural language generation tasks without any explicit task-specific knowledge or architecture of constituent parsing.

Copied to clipboard

Boxuan Lyu, Hidetaka Kamigaito, Kotaro Funakoshi, Manabu Okumura

Context-aware Neural Machine Translation with Coreference Information (D19-65)

Challenge:	Maximum a posteriori decoding aims to maximize the estimated posterior probability, but high estimated probability does not always lead to high translation quality.
Approach:	They propose a method that seeks hypotheses with the highest expected utility by using quasi-sources as “support hypothese . they propose sMBR decoding which utilizes a reference-free quality estimation metric as the utility function.
Outcome:	The proposed approach outperforms QE reranking and the standard MBR decoding.

Copied to clipboard

Takumi Ohtani, Hidetaka Kamigaito, Masaaki Nagata, Manabu Okumura

HalluCitation Matters: Revealing the Impact of Hallucinated References with 300 Hallucinated Papers in ACL Conferences (2026.acl-long)

Challenge:	Existing models for translating a sentence in a text do not consider coreference relations provided within the text.
Approach:	They propose a graph-based encoder which can consider coreference relations provided within the text explicitly.
Outcome:	The proposed model improves on the previous approach by 0.9 points on the BLEU score . the graph-based encoder can handle a longer text well, compared with the previous model .

Copied to clipboard

Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

Reasoning

Challenge:	citations that do not correspond to any existing work are a serious concern to scientific reliability and credibility.
Approach:	They analyze papers published at ACL, NAACL, and EMNLP in 2024 and 2025 . they identify 300 papers with at least one HalluCitation, most of which were published in 2025.
Outcome:	The authors analyze papers published at ACL, NAACL, and EMNLP in 2024 and 2025 . they find that nearly 300 papers contain at least one HalluCitation, most of which were published in 2025.

HLU: Human Vs LLM Generated Text Detection Dataset for Urdu at Multiple Granularities (2025.coling-main)

Copied to clipboard

Iqra Ali, Jesse Atuhurra, Hidetaka Kamigaito, Taro Watanabe

Neural text normalization leveraging similarities of strings and sounds (2020.coling-main)

Challenge:	Using large language models (LLMs) to generate human-like text has raised concerns about misuse, especially in low-resource languages like Urdu.
Approach:	They propose a dataset that contains documents, paragraphs, and sentences . they conducted human evaluations and automated evaluations .
Outcome:	The proposed dataset shows that distinguishing between human and machine-generated text is challenging for both humans and LLMs.

Copied to clipboard

Riku Kawamura, Tatsuya Aoki, Hidetaka Kamigaito, Hiroya Takamura, Manabu Okumura

Syntactic Text Processing

Challenge:	Existing methods that ignore the similarities of word strings and sounds do not account for these features.
Approach:	They propose a neural model that considers the similarities of both word strings and sounds, and a model that takes only the similarity of word strings or of sounds as a baseline.
Outcome:	The proposed models outperformed a baseline model and achieved state-of-the-art results on WNUT-2015.

Efficient Nearest Neighbor based Uncertainty Estimation for Natural Language Processing Tasks (2025.findings-naacl)

Copied to clipboard

Wataru Hashimoto, Hidetaka Kamigaito, Taro Watanabe

Green & Sustainable NLP

Challenge:	Existing methods for uncertainty estimation are inadequate for safety-critical applications.
Approach:	They propose a method that uses the distances from neighbors and the ratio of labels in neighbors to estimate uncertainty.
Outcome:	The proposed method outperforms baseline and density-based methods in calibration and uncertainty metrics.

AdParaphrase v2.0: Generating Attractive Ad Texts Using a Preference-Annotated Paraphrase Dataset (2025.findings-acl)

Copied to clipboard

Soichiro Murakami, Peinan Zhang, Hidetaka Kamigaito, Hiroya Takamura, Manabu Okumura

Considering Nested Tree Structure in Sentence Extractive Summarization with Pre-trained Transformer (2021.emnlp-main)

Challenge:	Identifying factors that make ad text attractive is essential for advertising success . identifying the linguistic factors presents a significant challenge because of the intricate interplay between the semantic content and its linguistic expression.
Approach:	They propose to use a dataset for ad text paraphrasing that contains human preference data to enable analysis of linguistic factors.
Outcome:	The proposed dataset is 20 times larger than v1.0 and contains 16,460 pairs of ad text paraphrase pairs . it shows that human preference and ade- t attractiveness are related .

Copied to clipboard

Jingun Kwon, Naoki Kobayashi, Hidetaka Kamigaito, Manabu Okumura

Measuring Watermarking under Jailbreaking: ASR Inflation and Goal-Compliance Mismatch (2026.findings-acl)

Challenge:	Sentence extractive summarization shortens a document by selecting sentences for a summary while preserving its important contents.
Approach:	They propose a nested tree-based extractive summarization model on RoBERTa that uses syntactic and discourse trees to represent sentences in a given document.
Outcome:	The proposed model outperforms baseline models on the CNN/DailyMail dataset and achieves significantly better scores than the baseline models in terms of coherence and comparable scores to the state-of-the-art models.

Copied to clipboard

Sungwoo Han, Sangjun Moon, Jingun Kwon, Hidetaka Kamigaito, Manabu Okumura

Long-Tail Crisis in Nearest Neighbor Language Models (2025.findings-naacl)

Challenge:	Recent work studies watermarking under benign prompts, but its behavior under jailbreaking prompts remains underexplored.
Approach:	They evaluate six methods on four LLMs using two jailbreak benchmarks and three settings: Static, AutoDAN, and DSN.
Outcome:	The proposed methods inflate judge-based attack success rate under jailbreaking, but not harmful-goal compliance.

Copied to clipboard

Yuto Nishida, Makoto Morishita, Hiroyuki Deguchi, Hidetaka Kamigaito, Taro Watanabe

XQ-MEval: A Dataset with Cross-lingual Parallel Quality for Benchmarking Translation Metrics (2026.findings-acl)

Challenge:	Prior studies have shown that kNN-LM can retrieve long-tail contexts, leaving the model’s performance underexplored in estimating the probabilities of long-tailed target tokens.
Approach:	They investigate the behavior of kNN-LM on low-frequency tokens, examining prediction probability, retrieval accuracy, and token distribution in the datastore.
Outcome:	The proposed model improves the perplexity of given text by directly accessing a large datastore built from any text data during inference.

Copied to clipboard

Jingxuan Liu, Zhi Qu, Jin Tei, Hidetaka Kamigaito, Lemao Liu, Taro Watanabe

Automatic Pyramid Evaluation Exploiting EDU-based Extractive Reference Summaries (D18-1)

Challenge:	averaging metric scores across languages is suspicious since translations of equal quality receive different scores across language.
Approach:	They propose a semi-automatically built dataset to benchmark translation metrics using MQM-defined errors and a normalization strategy to mitigate cross-lingual scoring bias.
Outcome:	The proposed model shows that translation metrics suffer from cross-lingual scoring bias . the proposed model is based on a semi-automatically built dataset covering nine translation directions .

Copied to clipboard

Tsutomu Hirao, Hidetaka Kamigaito, Masaaki Nagata

Challenge:	Existing methods for evaluating content are not accurate because they only confirm if the summary contains small textual fragments.
Approach:	They propose to transform human-made reference summaries into extractive reference sums and weight them using elementary discourse units.
Outcome:	The proposed method strongly correlates with manual evaluations on DUC and TAC data sets.

An Empirical Study of Generating Texts for Search Engine Advertising (2021.naacl-industry)

Copied to clipboard

Hidetaka Kamigaito, Peinan Zhang, Hiroya Takamura, Manabu Okumura

Information Retrieval

Challenge:	Existing studies on neural language generation have not evaluated the effect of generated ads with actual serving included because it requires a large amount of training data and a particular environment.
Approach:	They propose to integrate a reinforcement learning framework into an end-to-end sequence-tosequence (Seq2S) model and demonstrate how to improve the ads’ impact, deploy models to a product, and evaluate the generated ads.
Outcome:	The proposed method improves the ads’ impact, deploys the models to a product, and evaluates the generated ads.

Diagnosing Vision Language Models’ Perception by Leveraging Human Methods for Color Vision Deficiencies (2026.eacl-long)

Copied to clipboard

Kazuki Hayashi, Shintaro Ozaki, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

Dictionaries to the Rescue: Cross-Lingual Vocabulary Transfer for Low-Resource Languages Using Bilingual Dictionaries (2025.findings-acl)

Challenge:	Large-scale Vision-Language Models (LVLMs) are being deployed in real-world settings that require visual inference.
Approach:	They evaluate LVLMs' ability to account for variation in color perception using the Ishihara Test.
Outcome:	The proposed models fail to reproduce the perceptual outcomes experienced by affected individuals and default to normative color perception.

Copied to clipboard

Haruki Sakajo, Yusuke Ide, Justin Vasselli, Yusuke Sakai, Yingtao Tian, Hidetaka Kamigaito, Taro Watanabe

Simul-MuST-C: Simultaneous Multilingual Speech Translation Corpus Using Large Language Model (2024.emnlp-main)

Challenge:	Existing approaches to cross-lingual vocabulary transfer face challenges when dealing with low-resource languages.
Approach:	They propose a dictionary-based crosslingual vocabulary transfer method that leverages bilingual dictionaries, which are available for many languages thanks to descriptive linguists.
Outcome:	The proposed method outperforms existing methods for low-resource languages.

Copied to clipboard

Mana Makinae, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

Hierarchical Trivia Fact Extraction from Wikipedia Articles (2020.coling-main)

Challenge:	Simultaneous speech translation (SiST) begins translating before the entire source input is received.
Approach:	They propose a dataset that rearranges sentences into segmented monotonic data for simultaneous speech translation using the Large Language Model.
Outcome:	The proposed dataset improves quality and latency in siST translations by rearranging sentences into segmented monotonic data.

Copied to clipboard

Jingun Kwon, Hidetaka Kamigaito, Young-In Song, Manabu Okumura

Improving Neural RST Parsing Model with Silver Agreement Subtrees (2021.naacl-main)

Challenge:	Existing methods for extracting trivia facts for Wikipedia categories are not efficient . a trivia fact is an interesting fact that is unusual, unexpected, or unique .
Approach:	They propose an unsupervised algorithm that automatically mines trivia facts for a given entity . they propose to target at a single Wikipedia article and leverage its hierarchical structure .
Outcome:	The proposed algorithm outperforms existing methods and is 100 times faster than existing methods.

Copied to clipboard

Naoki Kobayashi, Tsutomu Hirao, Hidetaka Kamigaito, Manabu Okumura, Masaaki Nagata

Challenge:	Existing methods for Rhetorical Structure Theory (RST) parsing use supervised learning, but the RST-DT is small due to the costly annotation of RST trees.
Approach:	They propose to use silver data to improve RST parsing models by using annotated silver data.
Outcome:	The proposed method achieves the best micro-F1 scores for Nuclearity and Relation at 75.0 and 63.2 . it also achieves a remarkable gain in relation score against the previous state-of-the-art parser.

Towards Cross-Lingual Explanation of Artwork in Large-scale Vision Language Models (2025.findings-naacl)

Copied to clipboard

Shintaro Ozaki, Kazuki Hayashi, Yusuke Sakai, Hidetaka Kamigaito, Katsuhiko Hayashi, Taro Watanabe

Towards Table-to-Text Generation with Numerical Reasoning (2021.acl-long)

Challenge:	LVLMs are increasingly capable of responding in multiple languages . however, there is a lack of evaluation tools for LVLs that handle multiple languages.
Approach:	They used an extended dataset in multiple languages to evaluate LVLMs' ability to generate explanations in multiple language combinations.
Outcome:	The proposed dataset in multiple languages evaluates LVLMs' ability to generate explanations in other languages.

Copied to clipboard

Lya Hulliyyatus Suadaa, Hidetaka Kamigaito, Kotaro Funakoshi, Manabu Okumura, Hiroya Takamura

VLURes: Benchmarking Long-Text Grounding and Cross-Lingual Robustness in Vision Language Models (2026.findings-acl)

Challenge:	Recent studies have shown improvement in generating descriptive text from structured data.
Approach:	They propose a framework for numerical table-to-text generation based on numerical reasoning . they use a pre-trained model and a copy mechanism to fine-tune the models to produce fluent text .
Outcome:	The proposed framework lacks fidelity to the table contents and is based on a pre-trained model and a copy mechanism.

Copied to clipboard

Jesse Atuhurra, Iqra Ali, Tomoya Iwakura, Hidetaka Kamigaito, Tatsuya Hiraoka

Simultaneous Interpretation Corpus Construction by Large Language Models in Distant Language Pair (2024.emnlp-main)

Challenge:	*VLURes provides a practical testbed for long-text grounding and multilingual robustness in web-realistic agent settings.
Approach:	They propose a multilingual benchmark for evaluating vision-language models under long-text grounding.
Outcome:	*VLURes provides a testbed for long-text grounding and multilingual robustness in web-realistic agent settings.

Copied to clipboard

Yusuke Sakai, Mana Makinae, Hidetaka Kamigaito, Taro Watanabe

A New Surprise Measure for Extracting Interesting Relationships between Persons (2021.eacl-demos)

Challenge:	Existing siMT corpora are limited due to high costs and limited annotator capabilities.
Approach:	They propose a method to convert ST corpora into interpretation-style corpors by fine-tuning models with Large Language Models.
Outcome:	The proposed method reduces latency while achieving better quality compared to other models.

Copied to clipboard

Hidetaka Kamigaito, Jingun Kwon, Young-In Song, Manabu Okumura

Challenge:	Interesting facts are useful information for a variety of important tasks.
Approach:	They propose a method that extracts all personal relationships from dependency trees and calculates surprise scores for distributed representations of the extracted relationships in an unsupervised manner.
Outcome:	The proposed method extracts all personal relationships from dependency trees for the texts and calculates surprise scores for distributed representations of the extracted relationships in an unsupervised manner.

Hierarchical Label Generation for Text Classification (2023.findings-eacl)

Copied to clipboard

Jingun Kwon, Hidetaka Kamigaito, Young-In Song, Manabu Okumura

Metric-Type Identification for Multi-Level Header Numerical Tables in Scientific Papers (2021.eacl-main)

Challenge:	None Hierarchical text classification (HTC) aims to assign the most relevant labels with their structure for a given document.
Approach:	They propose a method that captures the label hierarchy for real-world classification applications by using a taxonomic hierarchy.
Outcome:	The proposed method can generate unseen labels in subword level.

Copied to clipboard

Lya Hulliyyatus Suadaa, Hidetaka Kamigaito, Manabu Okumura, Hiroya Takamura

Exploring Intrinsic Language-specific Subspaces in Fine-tuning Multilingual Neural Machine Translation (2024.emnlp-main)

Challenge:	Numerical tables are used to present experimental results in scientific papers.
Approach:	They propose a task to extract metric-types from multi-level header numerical tables . they propose two joint-learning neural classification and generation schemes .
Outcome:	The proposed models handle in-header and out-of-headers metric-type identification problems.

Copied to clipboard

Zhe Cao, Zhi Qu, Hidetaka Kamigaito, Taro Watanabe

Fusing Label Embedding into BERT: An Efficient Improvement for Text Classification (2021.findings-acl)

Challenge:	Multilingual neural machine translation models support fine-tuning hundreds of languages simultaneously.
Approach:	They propose to fine-tune a language in its intrinsic subspace with a tiny fraction of entire parameters.
Outcome:	The proposed methods outperform full-parameter fine-tuning up to 2.25 spBLEU scores and reduce trainable parameters to 0.4% for high and medium-resource languages and 1.6% for low-resourced ones.

Copied to clipboard

Yijin Xiong, Yukun Feng, Hao Wu, Hidetaka Kamigaito, Manabu Okumura

Generating Repetitions with Appropriate Repeated Words (2022.naacl-main)

Challenge:	Existing methods to improve text classification performance of pre-trained models have been used to improve their performance.
Approach:	They propose a method for improving BERT's performance by using a label embedding technique while keeping almost the same computational cost.
Outcome:	The proposed method improves BERT's performance on six text classification benchmark datasets while keeping almost the same computational cost.

Copied to clipboard

Toshiki Kawamoto, Hidetaka Kamigaito, Kotaro Funakoshi, Manabu Okumura

Challenge:	Existing studies have focused on general response generation with neural network-based approaches, but none have addressed specific types of repetitions.
Approach:	They propose a weighted label smoothing method for explicitly learning which words to repeat during fine-tuning and a repetition scoring method that can output more appropriate repetitions during decoding.
Outcome:	The proposed method outperforms baselines in automatic and human evaluations on a pre-trained language model for generating repetitions.

InstructCMP: Length Control in Sentence Compression through Instruction-based Large Language Models (2024.findings-acl)

Copied to clipboard

Juseon-Do Juseon-Do, Hidetaka Kamigaito, Manabu Okumura, Jingun Kwon

Joint Learning-based Heterogeneous Graph Attention Network for Timeline Summarization (2022.naacl-main)

Challenge:	Existing sentences do not consider the length constraints in extractive summarization because of their limited model abilities.
Approach:	They propose an approach that incorporates length constraints without model modifications into sentences . they use traditional sentence compression datasets to transform them into instruction format .
Outcome:	The proposed method can consider the length constraint through instructions without model modifications.

Copied to clipboard

Jingyi You, Dongyuan Li, Hidetaka Kamigaito, Kotaro Funakoshi, Manabu Okumura

Progressive Visual Refinement for Multi-modal Summarization (2026.eacl-short)

Challenge:	Existing studies on timeline summarization ignore the information interaction between sentences and dates, and combine them as two separate tasks.
Approach:	They propose a joint learning-based heterogeneous graph attention network for timeline summarization (HeterTls) they combine date selection and event detection into a unified framework to improve extraction accuracy .
Outcome:	The proposed model outperforms state-of-the-art models on four datasets . it significantly outperformed the baseline models on ROUGE scores and date selection metrics .

Copied to clipboard

Ye Xiong, Hidetaka Kamigaito, Soichiro Murakami, Peinan Zhang, Hiroya Takamura, Manabu Okumura

Considering Length Diversity in Retrieval-Augmented Summarization (2025.findings-naacl)

Challenge:	Multi-modal summarization (MMS) is a critical research area driven by the proliferation of multimedia content.
Approach:	They propose a patch-refined visual information network to exploit multimodal information . they propose combining visual information with textual information to generate concise summaries .
Outcome:	Extensive experiments on two public MMS datasets show the superiority of the proposed model.

Copied to clipboard

Juseon-Do Juseon-Do, Jaesung Hwang, Jingun Kwon, Hidetaka Kamigaito, Manabu Okumura

Revisiting Compositional Generalization Capability of Large Language Models Considering Instruction Following Ability (2025.acl-long)

Challenge:	Existing methods that require exhaustive exemplar-exemplar relevance comparisons do not consider summary lengths.
Approach:	They propose a Diverse Length-aware Maximal Marginal Relevance algorithm to better control summary lengths.
Outcome:	The proposed algorithm reduces the computational cost and memory consumption while maintaining the same level of informativeness.

Copied to clipboard

Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

Bidirectional Transformer Reranker for Grammatical Error Correction (2023.findings-acl)

Challenge:	generative large language models (LLMs) compose sentences that include all given concepts but must generate sentences that adhere to the specified order.
Approach:	They propose a benchmark to evaluate compositional generalization and instruction-following abilities of generative large language models (LLMs) based on ordered coverage, which allows simultaneous evaluation of both abilities.
Outcome:	The proposed benchmark evaluates compositional generalization and instruction-following abilities of LLMs.

Copied to clipboard

Ying Zhang, Hidetaka Kamigaito, Manabu Okumura

IRR: Image Review Ranking Framework for Evaluating Vision-Language Models (2025.coling-main)

Challenge:	Pre-trained seq2seq models suffer from a prediction bias due to their unidirectional decoding.
Approach:	They propose a bidirectional Transformer reranker that re-estimates the probability of each candidate sentence generated by pre-trained seq2seq models.
Outcome:	The proposed model improves on the original model and gives a 59.52 GLEU score on the JFLEG corpus.

Copied to clipboard

Kazuki Hayashi, Kazuma Onishi, Toma Suzuki, Yusuke Ide, Seiji Gobara, Shigeki Saito, Yusuke Sakai, Hidetaka Kamigaito, Katsuhiko Hayashi, Taro Watanabe

Visual Data in NLP

Challenge:	Large-scale vision language models excel at generating factual content, but their ability to rank images from multiple perspectives has not been explored.
Approach:	They propose a framework to evaluate large-scale vision-language models by measuring their ability to rank image texts from multiple perspectives.
Outcome:	The proposed evaluation framework measures how closely LVLMs' judgments align with human interpretations.

Does Pre-trained Language Model Actually Infer Unseen Links in Knowledge Graph Completion? (2024.naacl-long)

Copied to clipboard

Yusuke Sakai, Hidetaka Kamigaito, Katsuhiko Hayashi, Taro Watanabe

BQA: Body Language Question Answering Dataset for Video Large Language Models (2025.acl-short)

Challenge:	Knowledge Graph Completion (KGC) is a task that infers unseen relationships between entities . traditional embedding-based methods infer missing links using only training data . a pre-trained language model (PLM)-based KGC may be ineffective in practical applications .
Approach:	They propose to use knowledge Graph Completion (KGC) to infer unseen relationships . traditional embedding-based KGC methods infer missing links only from training data . they argue that pre-trained language models acquire inference abilities through pre-training .
Outcome:	The proposed method improves performance even though it does not use memorized knowledge.

Copied to clipboard

Shintaro Ozaki, Kazuki Hayashi, Miyu Oba, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

TableMBR: Minimum Bayes Risk Table Generation Based on Structural Consistency (2026.acl-srw)

Challenge:	a large part of human communication relies on nonverbal cues such as facial expressions, eye contact, and body language.
Approach:	They propose to validate whether video large language models can correctly interpret body language from short clips of body language.
Outcome:	The proposed model can correctly interpret emotions from short clips of body language.

Copied to clipboard

Daiki Yoshida, Hiroyuki Deguchi, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

Towards Artwork Explanation in Large-scale Vision Language Models (2024.acl-short)

Challenge:	Experimental results show TableMBR outperforms the baseline, achieving relative improvements of up to 15% in F1 on Rotowire and 23% in accuracy on LiveSum.
Approach:	They propose a text-to-table task that generates structured data from unstructured text . they propose 'tableMBR' that maintains structural consistency through minimum Bayes risk decoding .
Outcome:	The proposed method outperforms the baseline and achieves relative improvements in F1 and LiveSum.

Copied to clipboard

Kazuki Hayashi, Yusuke Sakai, Hidetaka Kamigaito, Katsuhiko Hayashi, Taro Watanabe

Explainability & Interpretability in NLP

Challenge:	Large-scale Vision-Language Models (LVLMs) output text from images and instructions, demonstrating advanced capabilities in text generation and comprehension.
Approach:	They propose to use artwork explanation generation task to quantitatively assess the understanding and utilization of artworks knowledge.
Outcome:	The proposed task evaluates the understanding and utilization of knowledge about artworks from images and titles and generates explanations using only images.

Monolingual Paraphrase Detection Corpus for Low Resource Pashto Language at Sentence Level (2024.lrec-main)

Copied to clipboard

Iqra Ali, Hidetaka Kamigaito, Taro Watanabe

Cross-lingual Contextualized Phrase Retrieval (2024.findings-emnlp)

Challenge:	Existing research on sentence-level paraphrase detection in Pashto has focused on English, but no work has been done on low-resource Pashtone.
Approach:	They propose to annotate sentences in Pashto to detect paraphrases . they will publicize a subset of 1,800 instances from their corpus, free from licensing issues.
Outcome:	The proposed corpus contains 6,727 sentences, encompassing 3,687 paraphrased and 3,040 non-paraphrased sentences.

Copied to clipboard

Huayang Li, Deng Cai, Zhi Qu, Qu Cui, Hidetaka Kamigaito, Lemao Liu, Taro Watanabe

Beyond Sampling: Self-Sorting for Long-Context Ranking (2026.findings-eacl)

Challenge:	Phrase-level dense retrieval has shown many appealing characteristics in downstream NLP tasks.
Approach:	They propose a task formulation of dense retrieval, cross-lingual contextualized phrase retrieval . they extract pairs of cross-linguistic phrases using word alignment information .
Outcome:	The proposed task formulation surpasses baselines on the phrase retrieval task and a downstream task, i.e., machine translation, and achieves top-1 accuracy 13 points higher.

Copied to clipboard

Juseon Do, Sungwoo Han, Jingun Kwon, Hidetaka Kamigaito, Katsuhiko Hayashi, Taro Watanabe

Challenge:	Large language models (LLMs) remain unstable on long-context ranking.
Approach:	They propose a method that fuses explicit within-list positions with implicit cross-list preferences to score entities and return a top-k set.
Outcome:	Experimental results show that large language models remain unstable on long-context ranking .

Table and Image Generation for Investigating Knowledge of Entities in Pre-trained Vision and Language Models (2023.acl-short)

Copied to clipboard

Hidetaka Kamigaito, Katsuhiko Hayashi, Taro Watanabe

Higher-Order Syntactic Attention Network for Longer Sentence Compression (N18-1)

Challenge:	Existing knowledge about entities acquired from natural language models is not retained in pre-trained vision & language models.
Approach:	They propose a task to verify how knowledge about entities acquired from natural language is retained in Vision & Language (V&L) models.
Outcome:	The proposed model forgets part of its entity knowledge by pre-training to improve image related tasks.

Copied to clipboard

Hidetaka Kamigaito, Katsuhiko Hayashi, Tsutomu Hirao, Masaaki Nagata

A Language Model-based Generative Classifier for Sentence-level Discourse Parsing (2021.emnlp-main)

Challenge:	Existing sentence compression methods do not handle syntactic features, causing performance degradation . et al. (2015) reported that the longer the input sentences are, the worse the performance becomes.
Approach:	They propose a higher-order syntactic attention network that handles higher-level dependency features as an attention distribution on LSTM hidden states.
Outcome:	The proposed method outperforms baseline methods on a Google sentence compression dataset.

Copied to clipboard

Ying Zhang, Hidetaka Kamigaito, Manabu Okumura

ConRAS: Contrastive In-context Learning Framework for Retrieval-Augmented Summarization (2026.findings-eacl)

Challenge:	Existing methods to consider textual coherence are limited in labeled data.
Approach:	They propose a language model-based generative classifier that uses labels as input and embeds labels into their representations.
Outcome:	The proposed classifier achieves state-of-the-art in discourse segmentation and relation F1 scores with gold boundaries and automatically segmented boundaries.

Copied to clipboard

Juseon Do, Sungwoo Han, Jingun Kwon, Hidetaka Kamigaito, Manabu Okumura

Low-Resource NLP

Challenge:	Despite progress in natural language processing, the potential of contrastive learning remains unexplored.
Approach:	They propose a framework that injects contrastive objectives into in-context learning-based retrieval-augmented summarization.
Outcome:	The proposed framework outperforms state-of-the-art retrieval-augmented methods on three summarization benchmarks showing that it can distinguish between positive and negative samples without parameter updates.

SinhalaMMLU: A Comprehensive Benchmark for Evaluating Multitask Language Understanding in Sinhala (2025.emnlp-main)

Copied to clipboard

Ashmari Pramodya, Nirasha Nelki, Heshan Shalinda, Chamila Liyanage, Yusuke Sakai, Randil Pushpananda, Ruvan Weerasinghe, Hidetaka Kamigaito, Taro Watanabe

Video Discourse Parsing and Its Application to Multimodal Summarization: A Dataset and Baseline Approaches (2024.findings-emnlp)

Challenge:	Large Language Models (LLMs) have been evaluated mostly on global or anglocentric subjects, often neglecting low-resource languages and culturally specific content.
Approach:	They evaluate 26 Large Language Models using a multiple-choice question answering benchmark for Sinhala.
Outcome:	The new benchmarks show that Claude 3.5 sonnet and GPT-4o achieve the highest average accuracies, but overall performance remains limited.

Copied to clipboard

Tsutomu Hirao, Naoki Kobayashi, Hidetaka Kamigaito, Manabu Okumura, Akisato Kimura

Simul-COMET: A Quality Metric for Simultaneous Interpretation in Distant Language Pair Considering Word Order Difference (2026.findings-acl)

Challenge:	Fig. 1 shows the video's story structure and event relationships in discourse parsing.
Approach:	They propose to construct an RST tree for a video to represent its storyline and illustrate the event relationships between events.
Outcome:	The proposed model outperforms two existing approaches to video RST parsing: the ‘parsing after captioning’ framework and parser using visual features.

Copied to clipboard

Kosuke Doi, Mana Makinae, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

Generating Diverse Translation with Perturbed kNN-MT (2024.eacl-srw)

Challenge:	Simultaneous interpretation (SI) uses segmenting of source speech into chunks and translating them in order.
Approach:	They propose a variation of COMET that measures monotonicity for simultaneous interpretation . they train Simul-COMET on offline translation data and show stronger alignment with evaluation scores .
Outcome:	The proposed model shows stronger alignment with evaluation scores provided by interpreters than COMET.

Copied to clipboard

Yuto Nishida, Makoto Morishita, Hidetaka Kamigaito, Taro Watanabe

AdParaphrase: Paraphrase Dataset for Analyzing Linguistic Features toward Generating Attractive Ad Texts (2025.findings-naacl)

Challenge:	Existing methods to generate multiple translation candidates do not address the overcorrection problem, which discourages the model from generating synonymous expressions and leans toward gold standards, reducing the diversity in the candidates.
Approach:	They propose to introduce perturbed k-nearest neighbor machine translation (kNN-MT) to generate more diverse translations.
Outcome:	The proposed methods significantly improve candidate diversity and control diversity by tuning the perturbation’s magnitude.

Copied to clipboard

Soichiro Murakami, Peinan Zhang, Hidetaka Kamigaito, Hiroya Takamura, Manabu Okumura

Translating Movie Subtitles by Large Language Models using Movie-meta Information (2025.acl-srw)

Challenge:	Effective linguistic choices that attract potential customers play crucial roles in advertising success.
Approach:	They propose to use a paraphrase dataset to explore linguistic features of ad texts that influence human preferences to maximize the potential success of advertisements.
Outcome:	The proposed model improves the attractiveness of ad texts by focusing on human preferences.

Copied to clipboard

Ashmari Pramodya, Yusuke Sakai, Justin Vasselli, Hidetaka Kamigaito, Taro Watanabe

Reliability of Distribution Predictions by LLMs: Insights from Counterintuitive Pseudo-Distributions (2025.naacl-srw)

Challenge:	Large language models (LLMs) have advanced natural language processing by understanding, generating, and manipulating texts.
Approach:	They propose to use movie subtitle prompts to improve translation accuracy by incorporating movie meta-information into the models.
Outcome:	The proposed prompts improve translation accuracy and reduce computational effort.

Copied to clipboard

Toma Suzuki, Ayuki Katayama, Seiji Gobara, Ryo Tsujimoto, Hibiki Nakatani, Kazuki Hayashi, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

Can we obtain significant success in RST discourse parsing by using Large Language Models? (2024.eacl-long)

Challenge:	Recent studies highlight the use of Large Language Models (LLMs) for predicting response distributions as a cost-effective survey method.
Approach:	They examine whether LLMs can rationally estimate distributions when presented with explanations that are against commonsense.
Outcome:	The proposed models can rationally estimate distributions when presented with explanations that are against commonsense, but smaller or less human-optimized models follow explanations uncritically, compared to larger models that resist counterintuitive explanations by leveraging their pretraining-acquired knowledge.

Copied to clipboard

Aru Maekawa, Tsutomu Hirao, Hidetaka Kamigaito, Manabu Okumura

mCSQA: Multilingual Commonsense Reasoning Dataset with Unified Creation Strategy by Language Models and Humans (2024.findings-acl)

Challenge:	Experimental results show that LLMs with tens of billion parameters can perform discourse parsing tasks.
Approach:	They employ Llama 2 and fine-tune it with QLoRA to achieve similar results . they show that LLMs with tens of billion parameters can perform a wide range of NLP tasks .
Outcome:	The proposed model performs better than existing models on three benchmark datasets.

Copied to clipboard

Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

Reasoning

Challenge:	Currently, multilingual datasets are created through translation, which cannot evaluate such language-specific aspects.
Approach:	They propose to curate a dataset for language-specific knowledge and commonsense . they propose to use multilingual commonsensiaq to leverage language models for a more efficient construction .
Outcome:	The proposed method reduces the creation cost by using multilingual LMs to create QAs . the proposed approach is based on the construction process of CSQA but with language models .

Enhancing Factuality through Consensus and Consistency in Summarization Using Minimum Bayes Risk Decoding (2026.findings-acl)

Copied to clipboard

Riza Setiawan Soetedjo, Yusuke Sakai, Hidetaka Kamigaito, Jingun Kwon, Manabu Okumura, Taro Watanabe

mbrs: A Library for Minimum Bayes Risk Decoding (2024.emnlp-demo)

Challenge:	Existing methods to improve summarization quality are limited to using source as guidance . reranking can be effective, but there are limitations, such as relying on reference-free metrics and rely on a single metric.
Approach:	They propose a model that reranks model-generated summaries by considering consistency to the source document and consensus among the other candidates.
Outcome:	The proposed system is competitive with existing methods, with human evaluations further confirming that it is superior.

Copied to clipboard

Hiroyuki Deguchi, Yusuke Sakai, Hidetaka Kamigaito, Taro Watanabe

Challenge:	Minimum Bayes risk (MBRS) decoding is a decision rule of text generation tasks that outperforms conventional maximum a posteriori (MAP) decoders by selecting high-quality outputs based on quality or preference rather than probability.
Approach:	They propose to use minimum bayes risk (MBRS) decoding to determine outputs based on quality rather than probability.
Outcome:	MBRS is an MIT-licensed open-source project with a focus on speed, reproducibility, and extensibility.

CoAM: Corpus of All-Type Multiword Expressions (2025.acl-long)

Copied to clipboard

Yusuke Ide, Joshua Tanner, Adam Nohejl, Jacob Hoffman, Justin Vasselli, Hidetaka Kamigaito, Taro Watanabe

Challenge:	Existing datasets for multiword expressions are inconsistently annotated, limited to a single type of MWE, or limited in size.
Approach:	They propose to use a new interface to generate MWE annotations for the first time in a dataset of MWE identification.
Outcome:	The proposed model outperforms existing models on the DiMSUM dataset.

LoCt-Instruct: An Automatic Pipeline for Constructing Datasets of Logical Continuous Instructions (2025.emnlp-main)

Copied to clipboard

Hongyu Sun, Yusuke Sakai, Haruki Sakajo, Shintaro Ozaki, Kazuki Hayashi, Hidetaka Kamigaito, Taro Watanabe

Reasoning

Challenge:	Existing instruction following datasets lack logical coherence across turns, narrow topical breadth and heavy manual effort.
Approach:	They propose a pipeline that leverages LLMs’ reasoning capabilities to assemble rich, topic-related single-instruction data into multi-turn dialogues and produce chains that are logically coherent, progressively deepen in content, and span diverse domains without fixed templates or extensive human annotation.
Outcome:	The proposed pipeline improves the performance of existing LLMs by integrating multiple topic-related data into multi-turn dialogues without fixed templates or extensive human annotation.

Generating Weather Comments from Meteorological Simulations (2021.eacl-main)

Copied to clipboard

Soichiro Murakami, Sora Tanaka, Masatsugu Hangyo, Hidetaka Kamigaito, Kotaro Funakoshi, Hiroya Takamura, Manabu Okumura