BLAST RADIUS

Open-weight & owned stack · Training tuning

Token-Level Data Selection Targets Safer LLM Fine-Tuning

Sep 12, 2026

An ICLR 2026 paper proposed selecting training data at token level to improve safety when fine-tuning language models.

Read the original at mlanthology.orgOpens the publisher's site in a new tab

More in Open-weight & owned stack