double-blind-review

Tag

Cards List
#double-blind-review

Large Language Models Threaten Double-blind Review

arXiv cs.CL · 2026-08-07 Cached

This paper demonstrates that large language models can effectively deanonymize authors of scientific papers from titles and abstracts alone, threatening the validity of double-blind peer review. The authors argue that stable patterns in problem framing and research focus act as latent conceptual signatures of authorship, necessitating a re-evaluation of anonymity practices in AI-augmented research ecosystems.

0 favorites 0 likes
← Back to home

Submit Feedback