publications

2024

1. NAACL

   M3T: A New Benchmark Dataset for Multi-Modal Document-Level Machine Translation

   Benjamin Hsu, Xiaoyu Liu, Huayang Li, Yoshinari Fujinuma, Maria Nadejde, Xing Niu, Ron Litman, Yair Kittenplon, and Raghavendra Pappagari

   In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers), Jun 2024

   PDF Code Data

2023

1. EMNLP Findings

   

   A Multi-Modal Multilingual Benchmark for Document Image Classification

   Yoshinari Fujinuma*, Siddharth Varia*, Nishant Sankaran, Srikar Appalaraju, Bonan Min, and Yogarshi Vyas

   In Proceedings of the Findings of Empirical Methods in Natural Language Processing (EMNLP), Jun 2023

   License: CC-BY-SA 4.0 for Wikipedia texts, CC-BY-4.0 for Eurlex documents

   PDF Data
2. EMNLP

   

   Comparing Biases and the Impact of Multilingual Training across Multiple Languages

   Sharon Levy, Neha Anna John, Ling Liu, Yogarshi Vyas, Jie Ma, Yoshinari Fujinuma, Miguel Ballesteros, Vittorio Castelli, and Dan Roth

   In Proceedings of the Empirical Methods in Natural Language Processing (EMNLP), Jun 2023

   PDF
3. ACL Findings

   

   Diable: Efficient Dialogue State Tracking as Operations on Tables

   Pietro Lesci, Yoshinari Fujinuma, Momchil Hardalov, Chao Shang, and Lluis Marquez

   In Proceedings of the Findings of Association for Computational Linguistics (ACL), Jun 2023

   PDF Code

2022

1. ACL

   

   Match the Script, Adapt if Multilingual: Analyzing the Effect of Multilingual Pretraining on Cross-lingual Transferability

   Yoshinari Fujinuma, Jordan Boyd-Graber, and Katharina Kann

   In Proceedings of the Association for Computational Linguistics (ACL), Jun 2022

   PDF Code Talk (English) Talk (Japanese) Poster Slides

2021

1. Workshop

   

   Semi-Supervised Joint Estimation of Word and Document Readability

   Yoshinari Fujinuma, and Masato Hagiwara

   In Proceedings of the Fifteenth Workshop on Graph-Based Methods for Natural Language Processing (TextGraphs-15), Jun 2021

   PDF Code Data

2020

1. ACL

   

   Why Overfitting Isn’t Always Bad: Retrofitting Cross-Lingual Word Embeddings to Dictionaries

   Mozhi Zhang*, Yoshinari Fujinuma*, Michael J. Paul, and Jordan Boyd-Graber

   In Proceedings of the Association for Computational Linguistics (ACL), Jun 2020

   PDF Code
2. AAAI

   Exploiting Cross-Lingual Subword Similarities in Low-Resource Document Classification

   Mozhi Zhang, Yoshinari Fujinuma, and Jordan Boyd-Graber

   In The Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI), Jun 2020

   PDF

2019

1. ACL

   

   A Resource-Free Evaluation Metric for Cross-Lingual Word Embeddings based on Graph Modularity

   Yoshinari Fujinuma, Jordan Boyd-Graber, and Michael J. Paul

   In Proceedings of the Association for Computational Linguistics (ACL), Jun 2019

   PDF Code Talk (English) Slides
2. PLOS ONE

   

   Zika discourse in the Americas: A multilingual topic analysis of Twitter

   Dasha Pruss, Yoshinari Fujinuma, Ashlynn R. Daughton, Michael J. Paul, Brad Arnot, Danielle Albers Szafir, and Jordan Boyd-Graber

   PLOS ONE, May 2019

   Abs PDF

   This work examines Twitter discussion surrounding the 2015 outbreak of Zika, a virus that is most often mild but has been associated with serious birth defects and neurological syndromes. We introduce and analyze a collection of 3.9 million tweets mentioning Zika geolocated to Noarth and South America, where the virus is most prevalent. Using a multilingual topic model, we automatically identify and extract the key topics of discussion across the dataset in English, Spanish, and Portuguese. We examine the variation in Twitter activity across time and location, finding that rises in activity tend to follow to major events, and geographic rates of Zika-related discussion are moderately correlated with Zika incidence (ρ = .398).

2018

1. Workshop

   Exploiting Cross-Lingual Subword Similarities in Low-Resource Document Classification

   Mozhi Zhang, Yoshinari Fujinuma, and Jordan Boyd-Graber

   In ACL Workshop on Deep Learning Approaches for Low-Resource Natural Language Processing, May 2018

2017

1. IJCNLP

   Substring Frequency Features for Segmentation of Japanese Katakana Words with Unlabeled Corpora

   Yoshinari Fujinuma, and Alvin Grissom II

   In Proceedings of the Eighth International Joint Conference on Natural Language Processing (IJCNLP), May 2017

   PDF Code

2015

1. PACLIC

   Distant-supervised Language Model for Detecting Emotional Upsurge on Twitter

   Yoshinari Fujinuma, Hikaru Yokono, Pascual Martı́nez-Gómez, and Akiko Aizawa

   In Proceedings of the 29th Pacific Asia Conference on Language, Information and Computation (PACLIC), May 2015

   PDF