Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements
Agentic ESOpt uses evolution strategies for scalable full-parameter fine-tuning of long-horizon LLM agents via trajectory-level reward-weighted updates and parameter-context co-evo…
Hugging Face · Daily Papers
·Zhi Zheng, Rongsheng Chen
·
·▲ 91 upvotes
Este artigo está em destaque na seleção diária de papers do Hugging Face, curada pela comunidade de pesquisa em IA.
Autores: Zhi Zheng, Rongsheng Chen, Yunpeng Ba, Zhenkun Wang, Yee Whye Teh, Wee Sun Lee
- 91 upvotes da comunidade
- Temas: evolution strategies, reinforcement learning, long-horizon agentic reasoning, credit assignment, full-parameter optimization, Agentic ESOpt
Resumo
Resumo original (em inglês), extraído do paper:
Agentic ESOpt uses evolution strategies for scalable full-parameter fine-tuning of long-horizon LLM agents via trajectory-level reward-weighted updates and parameter-context co-evolution.