Low-bit quantized LLM agents can recover much of their task success through closed-loop guidance from a full-precision teacher, followed by unassis...
论文概览
- arXiv ID:
2611.13410 - 发布机构 / 来源:Hugging Face Daily Papers
- 作者:Genming Cui, Xiangyu Li, Zixu Hao et al.
- 社区热度:32 Upvotes
核心摘要(Abstract / AI Summary)
Low-bit quantized LLM agents can recover much of their task success through closed-loop guidance from a full-precision teacher, followed by unassisted student training on corrected and reference turns.
