Search Results for author: Jun Suzuki

Found 79 papers, 30 papers with code

Scene-Text Aware Image and Text Retrieval with Dual-Encoder

no code implementations ACL 2022 Shumpei Miyawaki, Taku Hasegawa, Kyosuke Nishida, Takuma Kato, Jun Suzuki

We tackle the tasks of image and text retrieval using a dual-encoder model in which images and text are encoded independently.

Text Retrieval

LLM-jp: A Cross-organizational Project for the Research and Development of Fully Open Japanese LLMs

no code implementations4 Jul 2024 LLM-jp, :, Akiko Aizawa, Eiji Aramaki, Bowen Chen, Fei Cheng, Hiroyuki Deguchi, Rintaro Enomoto, Kazuki Fujii, Kensuke Fukumoto, Takuya Fukushima, Namgi Han, Yuto Harada, Chikara Hashimoto, Tatsuya Hiraoka, Shohei Hisada, Sosuke Hosokawa, Lu Jie, Keisuke Kamata, Teruhito Kanazawa, Hiroki Kanezashi, Hiroshi Kataoka, Satoru Katsumata, Daisuke Kawahara, Seiya Kawano, Atsushi Keyaki, Keisuke Kiryu, Hirokazu Kiyomaru, Takashi Kodama, Takahiro Kubo, Yohei Kuga, Ryoma Kumon, Shuhei Kurita, Sadao Kurohashi, Conglong Li, Taiki Maekawa, Hiroshi Matsuda, Yusuke Miyao, Kentaro Mizuki, Sakae Mizuki, Yugo Murawaki, Ryo Nakamura, Taishi Nakamura, Kouta Nakayama, Tomoka Nakazato, Takuro Niitsuma, Jiro Nishitoba, Yusuke Oda, Hayato Ogawa, Takumi Okamoto, Naoaki Okazaki, Yohei Oseki, Shintaro Ozaki, Koki Ryu, Rafal Rzepka, Keisuke Sakaguchi, Shota Sasaki, Satoshi Sekine, Kohei Suda, Saku Sugawara, Issa Sugiura, Hiroaki Sugiyama, Hisami Suzuki, Jun Suzuki, Toyotaro Suzumura, Kensuke Tachibana, Yu Takagi, Kyosuke Takami, Koichi Takeda, Masashi Takeshita, Masahiro Tanaka, Kenjiro Taura, Arseny Tolmachev, Nobuhiro Ueda, Zhen Wan, Shuntaro Yada, Sakiko Yahata, Yuya Yamamoto, Yusuke Yamauchi, Hitomi Yanaka, Rio Yokota, Koichiro Yoshino

This paper introduces LLM-jp, a cross-organizational project for the research and development of Japanese large language models (LLMs).

Detecting Response Generation Not Requiring Factual Judgment

no code implementations14 Jun 2024 Ryohei Kamei, Daiki Shiono, Reina Akama, Jun Suzuki

With the remarkable development of large language models (LLMs), ensuring the factuality of output has become a challenge.

Classification Response Generation

InstructDoc: A Dataset for Zero-Shot Generalization of Visual Document Understanding with Instructions

1 code implementation24 Jan 2024 Ryota Tanaka, Taichi Iki, Kyosuke Nishida, Kuniko Saito, Jun Suzuki

We study the problem of completing various visual document understanding (VDU) tasks, e. g., question answering and information extraction, on real-world documents through human-written instructions.

document understanding Question Answering +1

A Challenging Multimodal Video Summary: Simultaneously Extracting and Generating Keyframe-Caption Pairs from Video

1 code implementation4 Dec 2023 Keito Kudo, Haruki Nagasawa, Jun Suzuki, Nobuyuki Shimizu

This paper proposes a practical multimodal video summarization task setting and a dataset to train and evaluate the task.

Video Summarization

Refactoring Programs Using Large Language Models with Few-Shot Examples

no code implementations20 Nov 2023 Atsushi Shirafuji, Yusuke Oda, Jun Suzuki, Makoto Morishita, Yutaka Watanobe

A less complex and more straightforward program is a crucial factor that enhances its maintainability and makes writing secure and bug-free programs easier.

Language Modelling Large Language Model

Assessing Step-by-Step Reasoning against Lexical Negation: A Case Study on Syllogism

1 code implementation23 Oct 2023 Mengyu Ye, Tatsuki Kuribayashi, Jun Suzuki, Goro Kobayashi, Hiroaki Funayama

Large language models (LLMs) take advantage of step-by-step reasoning instructions, e. g., chain-of-thought (CoT) prompting.

Logical Reasoning Negation

Chat Translation Error Detection for Assisting Cross-lingual Communications

1 code implementation2 Aug 2023 Yunmeng Li, Jun Suzuki, Makoto Morishita, Kaori Abe, Ryoko Tokuhisa, Ana Brassard, Kentaro Inui

In this paper, we describe the development of a communication support system that detects erroneous translations to facilitate crosslingual communications due to the limitations of current machine chat translation methods.

Translation

Exploring the Robustness of Large Language Models for Solving Programming Problems

no code implementations26 Jun 2023 Atsushi Shirafuji, Yutaka Watanobe, Takumi Ito, Makoto Morishita, Yuki Nakamura, Yusuke Oda, Jun Suzuki

Our experimental results show that CodeGen and Codex are sensitive to the superficial modifications of problem descriptions and significantly impact code generation performance.

Code Generation

Bipartite-play Dialogue Collection for Practical Automatic Evaluation of Dialogue Systems

no code implementations19 Nov 2022 Shiki Sato, Yosuke Kishinami, Hiroaki Sugiyama, Reina Akama, Ryoko Tokuhisa, Jun Suzuki

Automation of dialogue system evaluation is a driving force for the efficient development of dialogue systems.

Domain Adaptation of Machine Translation with Crowdworkers

no code implementations28 Oct 2022 Makoto Morishita, Jun Suzuki, Masaaki Nagata

With the collected parallel data, we can quickly adapt a machine translation model to the target domain.

Domain Adaptation Machine Translation +1

Target-Guided Open-Domain Conversation Planning

1 code implementation COLING 2022 Yosuke Kishinami, Reina Akama, Shiki Sato, Ryoko Tokuhisa, Jun Suzuki, Kentaro Inui

Prior studies addressing target-oriented conversational tasks lack a crucial notion that has been intensively studied in the context of goal-oriented artificial intelligence agents, namely, planning.

Retrieval

N-best Response-based Analysis of Contradiction-awareness in Neural Response Generation Models

1 code implementation SIGDIAL (ACL) 2022 Shiki Sato, Reina Akama, Hiroki Ouchi, Ryoko Tokuhisa, Jun Suzuki, Kentaro Inui

In this scenario, the quality of the n-best list considerably affects the occurrence of contradictions because the final response is chosen from this n-best list.

Response Generation

Balancing Cost and Quality: An Exploration of Human-in-the-loop Frameworks for Automated Short Answer Scoring

no code implementations16 Jun 2022 Hiroaki Funayama, Tasuku Sato, Yuichiroh Matsubayashi, Tomoya Mizumoto, Jun Suzuki, Kentaro Inui

Towards guaranteeing high-quality predictions, we present the first study of exploring the use of human-in-the-loop framework for minimizing the grading cost while guaranteeing the grading quality by allowing a SAS model to share the grading task with a human grader.

B2T Connection: Serving Stability and Performance in Deep Transformers

1 code implementation1 Jun 2022 Sho Takase, Shun Kiyono, Sosuke Kobayashi, Jun Suzuki

Recent Transformers tend to be Pre-LN because, in Post-LN with deep Transformers (e. g., those with ten or more layers), the training is often unstable, resulting in useless models.

Text Generation

Towards Automated Document Revision: Grammatical Error Correction, Fluency Edits, and Beyond

1 code implementation23 May 2022 Masato Mita, Keisuke Sakaguchi, Masato Hagiwara, Tomoya Mizumoto, Jun Suzuki, Kentaro Inui

Natural language processing technology has rapidly improved automated grammatical error correction tasks, and the community begins to explore document-level revision as one of the next challenges.

Grammatical Error Correction Language Modelling +1

JParaCrawl v3.0: A Large-scale English-Japanese Parallel Corpus

no code implementations LREC 2022 Makoto Morishita, Katsuki Chousa, Jun Suzuki, Masaaki Nagata

Most current machine translation models are mainly trained with parallel corpora, and their translation accuracy largely depends on the quality and quantity of the corpora.

Machine Translation Sentence +1

SHAPE: Shifted Absolute Position Embedding for Transformers

1 code implementation13 Sep 2021 Shun Kiyono, Sosuke Kobayashi, Jun Suzuki, Kentaro Inui

Position representation is crucial for building position-aware representations in Transformers.

Position

Context-aware Neural Machine Translation with Mini-batch Embedding

1 code implementation EACL 2021 Makoto Morishita, Jun Suzuki, Tomoharu Iwata, Masaaki Nagata

It is crucial to provide an inter-sentence context in Neural Machine Translation (NMT) models for higher-quality translation.

Machine Translation NMT +2

An Investigation Between Schema Linking and Text-to-SQL Performance

1 code implementation3 Feb 2021 Yasufumi Taniguchi, Hiroki Nakayama, Kubo Takahiro, Jun Suzuki

Text-to-SQL is a crucial task toward developing methods for understanding natural language by computers.

Text-To-SQL

Quantum-state estimation problem via optimal design of experiments

no code implementations22 Dec 2020 Jun Suzuki

We then extend a classical result in the design problem, the Kiefer-Wolfowitz theorem, to a qubit system showing the D-optimal design is equivalent to a certain type of the A-optimal design.

Quantum Physics

Efficient Estimation of Influence of a Training Instance

no code implementations EMNLP (sustainlp) 2020 Sosuke Kobayashi, Sho Yokoi, Jun Suzuki, Kentaro Inui

Understanding the influence of a training instance on a neural network model leads to improving interpretability.

Langsmith: An Interactive Academic Text Revision System

no code implementations EMNLP 2020 Takumi Ito, Tatsuki Kuribayashi, Masatoshi Hidaka, Jun Suzuki, Kentaro Inui

Despite the current diversity and inclusion initiatives in the academic community, researchers with a non-native command of English still face significant obstacles when writing papers in English.

Diversity

Encoder-Decoder Models Can Benefit from Pre-trained Masked Language Models in Grammatical Error Correction

1 code implementation ACL 2020 Masahiro Kaneko, Masato Mita, Shun Kiyono, Jun Suzuki, Kentaro Inui

The answer to this question is not as straightforward as one might expect because the previous common methods for incorporating a MLM into an EncDec model have potential drawbacks when applied to GEC.

Decoder Grammatical Error Correction +1

Single Model Ensemble using Pseudo-Tags and Distinct Vectors

no code implementations ACL 2020 Ryosuke Kuwabara, Jun Suzuki, Hideki Nakayama

Model ensemble techniques often increase task performance in neural networks; however, they require increased time, memory, and management effort.

Management text-classification +1

Word Rotator's Distance

1 code implementation EMNLP 2020 Sho Yokoi, Ryo Takahashi, Reina Akama, Jun Suzuki, Kentaro Inui

Accordingly, we propose a method that first decouples word vectors into their norm and direction, and then computes alignment-based similarity using earth mover's distance (i. e., optimal transport cost), which we refer to as word rotator's distance.

Semantic Similarity Semantic Textual Similarity +3

Evaluating Dialogue Generation Systems via Response Selection

1 code implementation ACL 2020 Shiki Sato, Reina Akama, Hiroki Ouchi, Jun Suzuki, Kentaro Inui

Existing automatic evaluation metrics for open-domain dialogue response generation systems correlate poorly with human evaluation.

Dialogue Generation Response Generation

JParaCrawl: A Large Scale Web-Based English-Japanese Parallel Corpus

no code implementations LREC 2020 Makoto Morishita, Jun Suzuki, Masaaki Nagata

We constructed a parallel corpus for English-Japanese, for which the amount of publicly available parallel corpora is still limited.

Machine Translation Sentence +1

NTT Neural Machine Translation Systems at WAT 2019

no code implementations WS 2019 Makoto Morishita, Jun Suzuki, Masaaki Nagata

In this paper, we describe our systems that were submitted to the translation shared tasks at WAT 2019.

Machine Translation Translation

An Empirical Study of Incorporating Pseudo Data into Grammatical Error Correction

1 code implementation IJCNLP 2019 Shun Kiyono, Jun Suzuki, Masato Mita, Tomoya Mizumoto, Kentaro Inui

The incorporation of pseudo data in the training of grammatical error correction models has been one of the main factors in improving the performance of such models.

Grammatical Error Correction

The AIP-Tohoku System at the BEA-2019 Shared Task

no code implementations WS 2019 Hiroki Asano, Masato Mita, Tomoya Mizumoto, Jun Suzuki

We introduce the AIP-Tohoku grammatical error correction (GEC) system for the BEA-2019 shared task in Track 1 (Restricted Track) and Track 2 (Unrestricted Track) using the same system architecture.

Grammatical Error Detection Sentence

An Empirical Study of Span Representations in Argumentation Structure Parsing

no code implementations ACL 2019 Tatsuki Kuribayashi, Hiroki Ouchi, Naoya Inoue, Paul Reisert, Toshinori Miyoshi, Jun Suzuki, Kentaro Inui

For several natural language processing (NLP) tasks, span representation design is attracting considerable attention as a promising new technique; a common basis for an effective design has been established.

Effective Adversarial Regularization for Neural Machine Translation

1 code implementation ACL 2019 Motoki Sato, Jun Suzuki, Shun Kiyono

A regularization technique based on adversarial perturbation, which was initially developed in the field of image processing, has been successfully applied to text classification tasks and has yielded attractive improvements.

Machine Translation NMT +3

Character n-gram Embeddings to Improve RNN Language Models

no code implementations13 Jun 2019 Sho Takase, Jun Suzuki, Masaaki Nagata

This paper proposes a novel Recurrent Neural Network (RNN) language model that takes advantage of character information.

Headline Generation Language Modelling +3

Subword-based Compact Reconstruction of Word Embeddings

1 code implementation NAACL 2019 Shota Sasaki, Jun Suzuki, Kentaro Inui

The idea of subword-based word embeddings has been proposed in the literature, mainly for solving the out-of-vocabulary (OOV) word problem observed in standard word-based word embeddings.

Word Embeddings

The Sally Smedley Hyperpartisan News Detector at SemEval-2019 Task 4

no code implementations SEMEVAL 2019 Kazuaki Hanawa, Shota Sasaki, Hiroki Ouchi, Jun Suzuki, Kentaro Inui

Our system achieved 80. 9{\%} accuracy on the test set for the formal run and got the 3rd place out of 42 teams.

Annotating with Pros and Cons of Technologies in Computer Science Papers

1 code implementation WS 2019 Hono Shirai, Naoya Inoue, Jun Suzuki, Kentaro Inui

Specifically, we show how to adapt the targeted sentiment analysis task for pros/cons extraction in computer science papers and conduct an annotation study.

Sentiment Analysis

Mixture of Expert/Imitator Networks: Scalable Semi-supervised Learning Framework

no code implementations13 Oct 2018 Shun Kiyono, Jun Suzuki, Kentaro Inui

We also demonstrate that our method has the more data, better performance property with promising scalability to the amount of unlabeled data.

General Classification text-classification +1

NTT's Neural Machine Translation Systems for WMT 2018

no code implementations WS 2018 Makoto Morishita, Jun Suzuki, Masaaki Nagata

This paper describes NTT{'}s neural machine translation systems submitted to the WMT 2018 English-German and German-English news translation tasks.

Machine Translation Re-Ranking +1

Pointwise HSIC: A Linear-Time Kernelized Co-occurrence Norm for Sparse Linguistic Expressions

no code implementations EMNLP 2018 Sho Yokoi, Sosuke Kobayashi, Kenji Fukumizu, Jun Suzuki, Kentaro Inui

As well as deriving PMI from mutual information, we derive this new measure from the Hilbert--Schmidt independence criterion (HSIC); thus, we call the new measure the pointwise HSIC (PHSIC).

Machine Translation Sentence +2

Direct Output Connection for a High-Rank Language Model

1 code implementation EMNLP 2018 Sho Takase, Jun Suzuki, Masaaki Nagata

This paper proposes a state-of-the-art recurrent neural network (RNN) language model that combines probability distributions computed not only from a final RNN layer but also from middle layers.

Constituency Parsing Headline Generation +4

Improving Neural Machine Translation by Incorporating Hierarchical Subword Features

no code implementations COLING 2018 Makoto Morishita, Jun Suzuki, Masaaki Nagata

We hypothesize that in the NMT model, the appropriate subword units for the following three modules (layers) can differ: (1) the encoder embedding layer, (2) the decoder embedding layer, and (3) the decoder output layer.

Decoder Machine Translation +2

Interpretable Adversarial Perturbation in Input Embedding Space for Text

2 code implementations8 May 2018 Motoki Sato, Jun Suzuki, Hiroyuki Shindo, Yuji Matsumoto

This paper restores interpretability to such methods by restricting the directions of perturbations toward the existing words in the input embedding space.

Sentence

Input-to-Output Gate to Improve RNN Language Models

1 code implementation IJCNLP 2017 Sho Takase, Jun Suzuki, Masaaki Nagata

This paper proposes a reinforcing method that refines the output layers of existing Recurrent Neural Network (RNN) language models.

Enumeration of Extractive Oracle Summaries

no code implementations EACL 2017 Tsutomu Hirao, Masaaki Nishino, Jun Suzuki, Masaaki Nagata

To analyze the limitations and the future directions of the extractive summarization paradigm, this paper proposes an Integer Linear Programming (ILP) formulation to obtain extractive oracle summaries in terms of ROUGE-N. We also propose an algorithm that enumerates all of the oracle summaries for a set of reference summaries to exploit F-measures that evaluate which system summaries contain how many sentences that are extracted as an oracle summary.

document understanding Extractive Summarization

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