Repetition Mismatch Paper Accepted to EMNLP2026

Kris Cao retweeted a post by Kevin Zhou stating that their paper titled “Repetition Mismatch: Why Data Mixture Experiments Don’t Scale and How to Fix Them” will appear at EMNLP2026. The message includes the conference hashtag and a location abbreviation. Cao is identified in the post as a Research Engineer at Cohere working on pretraining large language models, with earlier roles at Google DeepMind and a PhD from Cambridge. No additional details about the paper content or acceptance status appear in the visible text.