AINeutralarXiv – CS AI · 10h ago5/10
🧠
Trajectory Supervision for Continual Tool-Use Learning in LLMs
Researchers demonstrate that preserving API request/response trajectories during continual learning significantly improves tool-use performance in language models. Fine-tuning Llama 3.1 8B on sequential API domains shows trajectory supervision achieves 56.9% accuracy versus 39.2% without intermediate context, though at a 25.1% token cost increase.
🧠 Llama