When Synthetic Data Hurts: On Catastrophic Forgetting in Skill Retrieval for LLM Agents
This work presents a production skill router over 34,396 skills and a large-scale study of skill retrieval using limited real supervision and synthetic data, and evaluates several forgetting mitigation fine-tuning approaches inspired by continual learning.
S. Murtaza, Yi-Fan Nie, Utkarsh Soni et al.
· 2 citations