Autonomous cyber defense systems based on Deep Reinforcement Learning (DRL) have attracted significant research attention, yet remain evaluated almost exclusively against static, heuristic red agents, leaving their robustness against adaptive threats critically understudied. Meanwhile, recent advances in Reinforcement Learning with Verifiable Rewards (RLVR) have improved LLM reasoning, but their integration into cybersecurity remains elusive due to the absence of suitable benchmark environments and interaction datasets.
Causal diagnostic models must explain how conclusions follow from evidence because diagnoses guide repairs and treatments. Yet serious cases are scarce, records rarely contain reasoning paths, and data transfer poorly across configurations, complicating local deployment.
Many critical reasoning tasks, including clinical diagnosis, legal judgment, and industrial fault diagnosis, require step-dependent causal chains in which early errors propagate and correct conclusions can mask invalid reasoning. Although large language models perform well on such tasks, privacy, latency, and controllability motivate distillation into locally deployable models.
Self-consistency assumes the most frequent answer among sampled reasoning traces is the most reliable, but this can fail in causal reasoning: samples often repeat the same confounding error, and votes fragment across multiple valid answers, letting an invalid answer win despite a valid minority trace. We introduce CALVER (Causal Axiom-Level VERification), a training-free symbolic verifier that scores structured traces against Pearl's causal criteria, including -separation, backdoor adjustment, and intervention, and selects the highest-scoring candidate without consulting a reference answer.
Historical tool-use trajectories provide valuable experience for large language model (LLM) agents to plan and coordinate tool usage. Existing approaches directly construct tool-level graphs from these trajectories, but the resulting graphs remain tied to specific tools and are hard to generalize across tool sets.
On-policy distillation, in which a teacher corrects samples that the student itself generates, presupposes that the two models speak the same language: identical VAE latents, matching architectures, and a common timestep grid. We ask what happens when none of this holds, as when the strongest teacher available and the student one wishes to deploy come from different model families, and find that the standard recipes have no answer: teacher latents cannot serve as targets in a foreign coordinate system, per-pixel losses against a teacher that stochastically re-draws local detail degenerate into blur or divergence, and timestep indices lose their meaning across mismatched schedules.
arXiv:2608. 01418v1 Announce Type: cross Abstract: Autoregressive rollout generation is a major computational cost in reinforcement learning for large language models.
By Wenhao Zhang, Yibo Xie, Rui Wang, Jiahua Yang, Lei Jiang, Zibo Yang, Yawei Wang, Jiali Xu, jasperawang, Haoyang Long, Huan Xiong, alantzhao
arXiv:2608. 02034v1 Announce Type: new Abstract: Multi-step returns accelerate reward propagation in off-policy reinforcement learning, but couple the evaluation of each decision to the suboptimal logged actions that follow it, inducing a pessimistic bias that grows with the horizon.
By Abdelghani Ghanem, Mounir Ghogho
arXiv:2608. 01678v1 Announce Type: new Abstract: Existing skill generation methods largely rely on heuristics or pipeline-style consolidation, which must be specially designed for different evidence sources.
By Junhao Shen, Zhanqiu Zhang, Yiwen Guo, Hong Cheng
arXiv:2607. 18082v3 Announce Type: replace Abstract: Rubric-based RL has recently shown promise in improving LLMs on open-ended tasks.
By Mingxuan Xia, Yuhang Yang, Chao Ye, Shuai Zhu, Shenzhi Yang, Guangcheng Zhu, Yuhang Zhang, Cheng Peng, Haobo Wang, Siqing Wang
arXiv:2608. 02508v1 Announce Type: new Abstract: Learning-based memory systems for self-evolving LLM agents face two tightly coupled challenges.
By Yi Yang, Zhennan Chen, Yihong Zhuang, Tiehan Fan, Yinan Chen, Jian Li, Jian Yang, Ying Tai
arXiv:2605. 08696v4 Announce Type: replace-cross Abstract: Over the last two decades, language modeling has experienced a shift from the use of predominantly recurrent architectures that process tokens sequentially during training and inference to non-recurrent models that process sequence elements in parallel during training, which results in greater training efficiency and stability at the expense of lower inference throughput.
By Benjamin L. Badger
arXiv:2608. 02181v1 Announce Type: new Abstract: Proximal Policy Optimization (PPO) for large language models typically trains its critic by mean-squared-error (MSE) regression on scalar value targets.
By Zhijian Zhou, Long Li, Xuan Zhang, Zongkai Liu, Yulei Qin, Ke Li, Xing Sun, Xiaoyu Tan, Chao Qu, Yuan Qi
arXiv:2608. 01743v1 Announce Type: new Abstract: Reinforcement learning (RL) has become a central paradigm for large language model (LLM) post-training, but optimization toward new objectives can degrade capabilities already present in the base model.
By Li Wang, Xiaodong Lu, Xiaohan Wang, Jiajun Chai, Wei Lin, Tianhao Peng, Guojun Yin
arXiv:2510. 15127v3 Announce Type: replace-cross Abstract: Identifying the effects of mechanical ventilation (MV) protocols in critical care requires analyzing data from heterogeneous patient-ventilator systems in the clinical decision-making environment.
By David J. Albers, Tell D. Bennett, Jana de Wiljes, George Hripcsak, Bradford J. Smith, Peter D. Sottile, J. N. Stroh
arXiv:2502. 17055v5 Announce Type: replace Abstract: Training instability in modern deep learning systems is frequently triggered by rare but extreme gradient-norm spikes, which can induce oversized parameter updates, corrupt optimizer state, and lead to slow recovery or divergence.
By Tianjin Huang, Zhangyang Wang, Haotian Hu, Zhenyu Zhang, Gaojie Jin, Xiang Li, Li Shen, Jiaxing Shang, Tianlong Chen, Ke Li, Lu Liu, Qingsong Wen, Shiwei Liu
arXiv:2604. 10727v2 Announce Type: replace-cross Abstract: Classical information-theoretic learning bounds typically rely on KL mutual information and moment-generating-function (MGF) arguments, which are well matched to bounded or sub-Gaussian losses but can be ineffective when losses or rewards are heavy-tailed.
By Huiming Zhang, Binghan Li, Wan Tian, Qiang Sun
arXiv:2604. 06543v2 Announce Type: replace-cross Abstract: In this work, we demonstrate that reliable stochastic sampling is a fundamental yet unfulfilled requirement for Large Language Models (LLMs) operating as agents.
By Xiangming Gu, Soham De, Michalis Titsias, Larisa Markeeva, Petar Veli\v{c}kovi\'c, Razvan Pascanu
arXiv:2608. 01917v1 Announce Type: new Abstract: Discounted exponential utility provides a principled criterion for risk-sensitive sequential decision-making, but its nonlinear structure complicates reinforcement learning.
By Ankur Naskar, Vivek T A, Aditya Kumar, Gugan Thoppe, Prashanth L. A
arXiv:2608. 01879v1 Announce Type: new Abstract: Tabular data generation supports analysis and decision-making when target-domain data are scarce, yet collecting complete target samples is often costly.
By Zijian Shen, Taijie Chen, Bin Zhou, Ziyang Jiang, Jintao Ke