token level correction

onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correctionarxiv.org

onPanda is an interactive annotation tool that lets humans correct LLM outputs token‑by‑token, then resumes generation from the corrected prefix. In a controlled study it cut median annotation time by 52% and the authors release a token‑level correction dataset (Panda‑CVL) for on‑policy fine‑tuning.