Papers by Katsuki Chousa

9 papers
Incorporating Noisy Length Constraints into Transformer with Length-aware Positional Encodings (2020.coling-main)

Copied to clipboard

Challenge: Neural Machine Translation suffers from an under-translation problem due to limited modeling of output sequence lengths.
Approach: They propose a method to train a Transformer model using length constraints based on positional encoding.
Outcome: The proposed method outperforms a vanilla Transformer in an English-to-Japanese translation by 3.22 points . the noise injection improved robustness for length prediction errors, especially within the window size.
WikiSplit++: Easy Data Refinement for Split and Rephrase (2024.lrec-main)

Copied to clipboard

Challenge: Existing text simplification methods rely on encoder-decoder models to achieve this task.
Approach: They propose a text-to-text generation approach that applies encoder-decoder models to a large-scale dataset to improve Split and Rephrase.
Outcome: The proposed approach improves Split and Rephrase readability and performance on large datasets, but still suffers from hallucinations and under-splitting.
Automatic Evaluation of Language Generation Technology Based on Structure Alignment (2025.coling-main)

Copied to clipboard

Challenge: Existing methods for automatic evaluation ignore syntax of sentences despite its importance in determining meaning.
Approach: They propose an automatic evaluation metric that considers both the words in sentences and their syntactic structures.
Outcome: The proposed method is comparable to baselines from two NLP tasks.
One Single Hub Text Breaks CLIP: Identifying Vulnerabilities in Cross-Modal Encoders via Hubness (2026.acl-long)

Copied to clipboard

Challenge: et al., 2010) show that hub embeddings are close to many unrelated examples in high-dimensional embeddable spaces . cross-modal encoders that project different modalities into a shared space are useful for cross-module applications .
Approach: They propose a method for identifying the hub embedding and its corresponding hub text . they use images to evaluate cross-modal encoders that project different modalities into a shared space .
Outcome: The proposed method can identify a single hub embedding and its corresponding hub text . it achieves comparable or higher similarity scores than human-written reference captions in many images .
JaParaPat: A Large-Scale Japanese-English Parallel Patent Application Corpus (2024.lrec-main)

Copied to clipboard

Challenge: a recent study has demonstrated that patent translation accuracy improves as the amount of training data or the number of model parameters increases.
Approach: They construct a bilingual corpus of Japanese-English patent application data from 2000 to 2021 . they extracted 1.4M Japanese- English document pairs and extracted 350M sentence pairs .
Outcome: The proposed method improves translation accuracy by 20 bleu points . it is the first publicly available large-scale Japanese-English patent corpus .
SpanAlign: Sentence Alignment Method based on Cross-Language Span Prediction and ILP (2020.coling-main)

Copied to clipboard

Challenge: Existing methods for automatic sentence alignment assume monotonic alignments, but they can handle non-monotonic alignments.
Approach: They propose a method to automatically extract parallel sentences from noisy parallel documents by embeddings and encoding each source and target sentence.
Outcome: The proposed method improves translation accuracy by 4.1 BLEU scores on English-Japanese . it can predict spans in target document from sentences in source document .
JParaCrawl v3.0: A Large-scale English-Japanese Parallel Corpus (2022.lrec-1)

Copied to clipboard

Challenge: Existing parallel corpora for English-Japanese are limited, limiting the accuracy of machine translation models.
Approach: They propose a web-based English-Japanese parallel corpus with 21 million unique sentence pairs . this is more than twice as many as the previous corpus JParaCrawl v2.0 .
Outcome: The proposed corpus boosts the accuracy of machine translation models on various domains.
Hacking Neural Evaluation Metrics with Single Hub Text (2026.eacl-short)

Copied to clipboard

Challenge: Recent embedding-based neural text evaluation metrics are not reliable due to black-box nature of neural networks.
Approach: They propose to find a single adversarial text in the discrete space that is consistently evaluated as high-quality regardless of the test cases.
Outcome: The proposed method outperforms translations generated individually for each source sentence in English-to-Japanese and English- to-German translation tasks.
A Supervised Word Alignment Method based on Cross-Language Span Prediction using Multilingual BERT (2020.emnlp-main)

Copied to clipboard

Challenge: supervised word alignment tools such as GIZA++, MGIZA (Gao and Vogel, 2008) and FastAlign remain stagnant in terms of word alignment accuracy.
Approach: They propose a supervised word alignment method based on cross-language span prediction by formalizing a word alignment problem as a collection of independent predictions from a token in the source sentence to a span in the target sentence.
Outcome: The proposed method significantly outperforms previous supervised and unsupervised word alignment methods without any bitexts for pretraining.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations