In the legal domain, LLMs can improve reliability and interpretability by generating explicit reasoning traces. However, training LLMs to produce high-qualit…
机构:阿里
来源:arXiv 2611.15346 | AI4Papers 论文推荐平台