1
Verifiable Process Rewards for Agentic Reasoning
提出可验证过程奖励机制，让智能体推理更可信可解释，强化学习新思路。
arXiv:2605.10325v2 Announce Type: replace Abstract: Reinforcement learning from verifiable rewards (RLVR) has improved the reasoning abilities of larg…
2
LGMT: Logic-Grounded Metamorphic Testing for Evaluating the Reasoning Reliability of LLMs
用逻辑蜕变测试方法评估大模型推理可靠性，LGMT框架揭示幻觉与逻辑不一致陷阱
arXiv:2605.23965v1 Announce Type: new Abstract: Large Language Models (LLMs) achieve strong performance on logical reasoning benchmarks, yet their rel…