today local_bar

Chao-Chung Wu

Researcher, Appier

1 paper at NeurIPS 2025

OpenReview· Semantic Scholar· Google Scholar

Poster Session 5

Friday, December 5, 2025 · 11:00 AM → 2:00 PM

Exhibit Hall C,D,E

Mitigating Forgetting in LLM Fine-Tuning via Low-Perplexity Token Learning

#3504 · Chao-Chung Wu, Zhi Rui Tam, Chieh-Yen Lin, Yun-Nung Chen, Shao-Hua Sun, Hung-yi Lee

A simple masking technique to avoid LLM finetuning from degrading on general capabilities.