Sampling multiple solutions and returning the majority answer is among the most reliable ways to improve the reasoning accuracy of large language models without labels, and a growing family of methods converts this consensus signal into training supervision. However, existing approaches use consensus only in restricted forms: as a filter that selects solutions for fine-tuning, as a preference between answers, or as a scalar reward for reinforcement learning, discarding most of the information that the agreeing solutions contain.
LLM agents are increasingly evaluated on multi-week decision tasks in which the state that drives cost is never directly observed. On such tasks the final cost cannot say why an agent failed: it may have misread the world, or read it correctly and still failed to act (the knowing-doing gap).
Large Language Model (LLM) agents increasingly rely on external memory systems to accumulate experience across tasks. Yet nearly all existing approaches, from graph-structured memories to reflective insight stores, access memory through fixed, hand-designed heuristics.
Mobile crowdsensing (MC) recruits mobile users to perform sensing tasks using their smartphones, enabling large-scale applications such as traffic monitoring and environmental sensing. A fundamental challenge is online worker recruitment under uncertainty, where the platform must learn workers' sensing performance while operating with a limited budget.
While traditional graphics methods often synthesize 3D indoor scenes autoregressively or hierarchically, recent vision-language model (VLM)-based generators predominantly adopt a one-shot paradigm where the full layout is planned at once. This one-shot approach often requires global re-optimization or complete reconstruction during interactive editing (e.
Reinforcement learning for multi-turn search reasoning typically relies on terminal outcome rewards, which cannot distinguish useful, redundant, and harmful intermediate interactions. We propose LOTAPO , a self-generated process-supervision method based on backward leave-one-turn attribution.
Reinforcement learning for multi-turn search reasoning typically relies on terminal outcome rewards, which cannot distinguish useful, redundant, and harmful intermediate interactions. We propose LAPO, a self-generated process-supervision method based on backward leave-one-turn attribution.
arXiv:2607. 11948v1 Announce Type: new Abstract: Regulated financial institutions operating under data-residency rules need tenant-owned language models that can run inside the institution's perimeter.
By Thanh Luong Tuan
arXiv:2412. 04847v2 Announce Type: replace-cross Abstract: Reinforcement learning (RL) has demonstrated remarkable capabilities in training agents to solve complex tasks autonomously, such as mobile robots, UAVs/UGVs, and game-playing agents).
By Rachmad Vidya Wicaksana Putra, Avaneesh Devkota, Muhammad Shafique
arXiv:2607. 12687v1 Announce Type: cross Abstract: LLMs can perform language-based quantitative prediction from unstructured inputs, but remain susceptible to hallucinations and overconfident errors, making it critical to know not only what a model predicts, but when its predictions can be trusted.
By Mehak Dhaliwal, Rasta Tadayon, Andong Hua, Haewon Jeong, Yao Qin
arXiv:2607. 11953v1 Announce Type: new Abstract: Does a reinforcement-learning agent that earns high reward represent its task's latent state, or only a reward-correlated shortcut?
By Jim Allchin
arXiv:2607. 12962v1 Announce Type: cross Abstract: Frozen small code LLMs are deployed locally, yet the information guiding a retry after a failed attempt is still measured without placebo controls in the self-repair literature.
By Mehmet Iscan
arXiv:2607. 12626v1 Announce Type: cross Abstract: Closed-loop wall control learnt by multi-agent reinforcement learning can lower skin-friction drag in turbulent channels, but these gradient-based policies are trained on small periodic boxes and exhibit reduced performance when carried over to a larger domain.
By Giorgio Maria Cavallazzi, Miguel P\'erez Cuadrado, Alfredo Pinelli
arXiv:2509. 24372v3 Announce Type: replace-cross Abstract: Fine-tuning large language models (LLMs) for downstream tasks is an essential stage of modern AI deployment.
By Xin Qiu, Yulu Gan, Conor F. Hayes, Qiyao Liang, Yinggan Xu, Roberto Dailey, Elliot Meyerson, Babak Hodjat, Risto Miikkulainen
arXiv:2603. 05995v2 Announce Type: replace-cross Abstract: Off-road autonomous driving poses significant challenges such as navigating unmapped, variable terrain with uncertain and diverse dynamics.
By Zhouchonghao Wu, Raymond Song, Vedant Mundheda, Luis E. Navarro-Serment, Christof Schoenborn, Jeff Schneider
arXiv:2607. 11906v1 Announce Type: new Abstract: The development of decision-pretrained transformers, algorithm distillation, long-context meta-RL, and retrieval-augmented agents has renewed interest in in-context reinforcement learning (ICRL): the ability of a pretrained or fine-tuned decision model to infer latent task rules and improve future behavior from interaction context, without test-time parameter updates.
By A Run, Ziluo Ding
arXiv:2607. 12389v1 Announce Type: cross Abstract: We consider Bayesian bandit models and prove that Thompson sampling makes at most twice the expected number of mistakes (selections of a suboptimal arm) as any other policy.
By Mark Sellke, Gregory Valiant
arXiv:2601. 19624v3 Announce Type: replace-cross Abstract: Real-world reinforcement learning often faces environment drift, but most existing methods rely on static entropy coefficients/target entropy, causing over-exploration during stable periods and under-exploration after drift, and leaving unanswered the principled question of how exploration intensity should scale with drift magnitude.
By Tongxi Wang, Zhuoyang Xia, Xinran Chen, Shan Liu
arXiv:2011. 02565v2 Announce Type: replace-cross Abstract: Temporal abstraction allows reinforcement learning agents to represent knowledge and develop strategies over different temporal scales.
By Anand Kamat, Doina Precup
arXiv:2607. 11954v1 Announce Type: cross Abstract: Clinical code prediction maps unstructured discharge summaries to ICD-10-CM leaf codes in a large, sparse, and deeply hierarchical label space.
By Amritpal Singh, Sebastian Torres, Khawar Shakeel, Syed Ahmad Chan Bukhari