Search Results for author: Koichi Takeda

Found 20 papers, 4 papers with code

Leveraging Three Types of Embeddings from Masked Language Models in Idiom Token Classification

no code implementations *SEM (NAACL) 2022 Ryosuke Takahashi, Ryohei Sasano, Koichi Takeda

Recent research has shown that contextualized word embeddings derived from masked language models (MLMs) can give promising results for idiom token classification.

Classification token-classification +2

WikiSplit++: Easy Data Refinement for Split and Rephrase

1 code implementation13 Apr 2024 Hayato Tsukagoshi, Tsutomu Hirao, Makoto Morishita, Katsuki Chousa, Ryohei Sasano, Koichi Takeda

The task of Split and Rephrase, which splits a complex sentence into multiple simple sentences with the same meaning, improves readability and enhances the performance of downstream tasks in natural language processing (NLP).

Decoder Sentence +2

Verifying Claims About Metaphors with Large-Scale Automatic Metaphor Identification

no code implementations1 Apr 2024 Kotaro Aono, Ryohei Sasano, Koichi Takeda

There are several linguistic claims about situations where words are more likely to be used as metaphors.

Improving Sentence Embeddings with an Automatically Generated NLI Dataset

no code implementations23 Feb 2024 Soma Sato, Hayato Tsukagoshi, Ryohei Sasano, Koichi Takeda

Decoder-based large language models (LLMs) have shown high performance on many tasks in natural language processing.

Decoder Natural Language Inference +5

Japanese SimCSE Technical Report

1 code implementation30 Oct 2023 Hayato Tsukagoshi, Ryohei Sasano, Koichi Takeda

We report the development of Japanese SimCSE, Japanese sentence embedding models fine-tuned with SimCSE.

Sentence Sentence Embedding +1

Transformer-based Live Update Generation for Soccer Matches from Microblog Posts

no code implementations25 Oct 2023 Masashi Oshika, Kosuke Yamada, Ryohei Sasano, Koichi Takeda

It has been known to be difficult to generate adequate sports updates from a sequence of vast amounts of diverse live tweets, although the live sports viewing experience with tweets is gaining the popularity.

Language Modelling

Acquiring Frame Element Knowledge with Deep Metric Learning for Semantic Frame Induction

no code implementations23 May 2023 Kosuke Yamada, Ryohei Sasano, Koichi Takeda

The semantic frame induction tasks are defined as a clustering of words into the frames that they evoke, and a clustering of their arguments according to the frame element roles that they should fill.

Clustering Language Modelling +1

Sentence Representations via Gaussian Embedding

no code implementations22 May 2023 Shohei Yoda, Hayato Tsukagoshi, Ryohei Sasano, Koichi Takeda

Recent progress in sentence embedding, which represents the meaning of a sentence as a point in a vector space, has achieved high performance on tasks such as a semantic textual similarity (STS) task.

Contrastive Learning Natural Language Inference +5

Semantic Frame Induction with Deep Metric Learning

no code implementations27 Apr 2023 Kosuke Yamada, Ryohei Sasano, Koichi Takeda

Recent studies have demonstrated the usefulness of contextualized word embeddings in unsupervised semantic frame induction.

Metric Learning Word Embeddings

Cross-Modal Similarity-Based Curriculum Learning for Image Captioning

no code implementations14 Dec 2022 Hongkuan Zhang, Saku Sugawara, Akiko Aizawa, Lei Zhou, Ryohei Sasano, Koichi Takeda

Moreover, the higher model performance on difficult examples and unseen data also demonstrates the generalization ability.

Image Captioning Language Modelling

Transformer-based Lexically Constrained Headline Generation

1 code implementation EMNLP 2021 Kosuke Yamada, Yuta Hitomi, Hideaki Tamori, Ryohei Sasano, Naoaki Okazaki, Kentaro Inui, Koichi Takeda

We also consider a new headline generation strategy that takes advantage of the controllable generation order of Transformer.

Headline Generation

Verb Sense Clustering using Contextualized Word Representations for Semantic Frame Induction

no code implementations Findings (ACL) 2021 Kosuke Yamada, Ryohei Sasano, Koichi Takeda

Furthermore, we examine the extent to which the contextualized representation of a verb can estimate the number of frames that the verb can evoke.


Semantic Frame Induction using Masked Word Embeddings and Two-Step Clustering

no code implementations ACL 2021 Kosuke Yamada, Ryohei Sasano, Koichi Takeda

Recent studies on semantic frame induction show that relatively high performance has been achieved by using clustering-based methods with contextualized word embeddings.

Clustering Vocal Bursts Valence Prediction +1

DefSent: Sentence Embeddings using Definition Sentences

1 code implementation ACL 2021 Hayato Tsukagoshi, Ryohei Sasano, Koichi Takeda

However, these methods are only available for limited languages due to relying heavily on the large NLI datasets.

Natural Language Inference Sentence +3

Self-Guided Curriculum Learning for Neural Machine Translation

no code implementations ACL (IWSLT) 2021 Lei Zhou, Liang Ding, Kevin Duh, Shinji Watanabe, Ryohei Sasano, Koichi Takeda

In the field of machine learning, the well-trained model is assumed to be able to recover the training labels, i. e. the synthetic labels predicted by the model should be as close to the ground-truth labels as possible.

Machine Translation NMT +2

Zero-Shot Translation Quality Estimation with Explicit Cross-Lingual Patterns

no code implementations WMT (EMNLP) 2020 Lei Zhou, Liang Ding, Koichi Takeda

In response to this issue, we propose to expose explicit cross-lingual patterns, \textit{e. g.} word alignments and generation score, to our proposed zero-shot models.

Sentence Translation

Development of a Medical Incident Report Corpus with Intention and Factuality Annotation

no code implementations LREC 2020 Hongkuan Zhang, Ryohei Sasano, Koichi Takeda, Zoie Shui-Yee Wong

In this paper, we present our annotation scheme with respect to the definition of medication entities that we take into account, the method to annotate the relations between entities, and the details of the intention and factuality annotation.

Incorporating Textual Information on User Behavior for Personality Prediction

no code implementations ACL 2019 Kosuke Yamada, Ryohei Sasano, Koichi Takeda

Our experiments on the personality prediction of Twitter users show that the textual information of user behaviors is more useful than the co-occurrence information of the user behaviors.

Cannot find the paper you are looking for? You can Submit a new open access paper.