arXiv 论文速递

揭示思维链监控可被计划注入攻击绕过，对AI安全对齐与监控机制的有效性提出重要警示。

Chain-of-thought (CoT) monitoring is a safety strategy where the reasoning of a large language model "actor" is inspected by a "monitor" (often another language model) for signs of…

提出基于贝尔曼方程的RLVR新算法，提升大模型推理能力，对强化学习与LLM推理训练有直接推动作用。

Reinforcement learning with verifiable rewards (RLVR) improves the reasoning capabilities of large language models (LLMs). We introduce Bellman Policy Optimization (BPO), a critic-…

构建多智能体长周期数学与理论计算机研究测试平台，推动LLM在复杂科研任务中的可靠评估。

Language models can produce plausible short proofs, but may still be unreliable on long-horizon research problems, where progress depends on a sequence of uncertain and interdepend…

Skills extend an LLM agent beyond its parametric knowledge, and the gain they promise rests on picking the right one. Deployed harnesses route by preloading every skill's metadata …

When a video model generates physically incorrect motion, did it fail to learn the correct motion, or did it learn it but fail to use it? We show the latter: the correct motion rem…

Transformer representations evolve through learned additive transformations that either preserve their current direction or redirect it. We study this evolution as a functional geo…

As language models become more capable, long-term collaboration in learning, reasoning, and decision-making calls for a deeper understanding of the people they serve. Yet training …

Record-level differential privacy exposes a structural misalignment in personalized federated learning when client-specific variation is low-dimensional while training repeatedly r…

提出仓库级漏洞定位基准，弥补智能体安全分析评估空白，对AI驱动软件安全研究意义重大。

Language-model agents increasingly operate over complete software repositories, yet cybersecurity evaluations primarily measure whether they can detect, reproduce, or repair vulner…

Recurrent GNNs iterate message passing to convergence, and their logical characterizations to date rely on multi-set aggregation, graded (counting) logics, and halting or acceptanc…

Artificial intelligence presents firms with an unusual timing problem. The technology frontier is improving rapidly, implementation is partly irreversible, and organization-specifi…

提出信息空间可达性方法实现安全元强化学习，解决新任务适应中的安全约束问题，具应用价值。

Meta-reinforcement learning (meta-RL) enables agents to adapt to unseen tasks with limited experience. Despite its promise, the application of meta-RL in real-world tasks is hinder…