Showing cs.CLShow all
3 papers · 1 filter
cs.CL2026
One Single Hub Text Breaks CLIP: Identifying Vulnerabilities in Cross-Modal Encoders via Hubness
Hiroyuki Deguchi, Katsuki Chousa, Yusuke Sakai
The hubness problem, in which hub embeddings are close to many unrelated examples, occurs often in high-dimensional embedding spaces and may pose a practical threat for purposes su…
cs.CL2026
Hacking Neural Evaluation Metrics with Single Hub Text
Hiroyuki Deguchi, Katsuki Chousa, Yusuke Sakai
Strongly human-correlated evaluation metrics serve as an essential compass for the development and improvement of generation models and must be highly reliable and robust. Recent e…
cs.CL2025
JaParaPat: A Large-Scale Japanese-English Parallel Patent Application Corpus
Masaaki Nagata, Katsuki Chousa, Norihito Yasuda
We constructed JaParaPat (Japanese-English Parallel Patent Application Corpus), a bilingual corpus of more than 300 million Japanese-English sentence pairs from patent applications…