1
The Bidirectional Process Reward Model
提出双向过程奖励模型，突破传统单向奖励局限，提升语言模型推理与对齐性能。
arXiv:2508.01682v3 Announce Type: replace Abstract: Process Reward Models (PRMs), which assign fine-grained scores to intermediate reasoning steps wit…
2
Mental-R1: Aligning LLM Reasoning for Mental Health Assessment
新方法Mental-R1通过对齐LLM推理，显著提升心理健康评估的准确性与可靠性
arXiv:2606.13176v1 Announce Type: new Abstract: Mental health problems such as anxiety, depression, and suicide remain urgent global challenges, where…
3
R2IF: Aligning Reasoning with Decisions via Composite Rewards for Interpretable LLM Function Calling
提出R2IF框架，用复合奖励让大模型函数调用时推理与决策对齐，提升可解释性。
arXiv:2604.20316v2 Announce Type: replace Abstract: Function calling empowers large language models (LLMs) to interface with external tools, yet exist…
4
Harnesses for Inference-Time Alignment over Execution Trajectories
探索推理时对齐与执行轨迹的新方法，为AI系统行为优化提供理论框架。
arXiv:2605.21516v1 Announce Type: cross Abstract: Harness engineering has emerged as an important inference-time technique for large language model (L…
5
CriterAlign: Criterion-Centric Rationale Alignment for Code Preference Judging
基于标准中心的推理对齐新范式，显著提升大模型代码偏好判断的准确性与可解释性。
arXiv:2605.19665v1 Announce Type: cross Abstract: Pairwise human preference prediction is central to evaluating code-generation systems, where quality…
6
ReAlign: Generalizable Image Forgery Detection via Reasoning-Aligned Representation
提出ReAlign方法，通过推理对齐表征实现高泛化性的AI生成图像伪造检测
arXiv:2605.16080v1 Announce Type: new Abstract: The rise of AI-generated images (AIGIs) poses growing challenges for digital authenticity, prompting t…