Latest Reinforcement Learning Research Papers

The newest Reinforcement Learning papers from across the field — arXiv, NeurIPS, CVPR, Nature, and more — refreshed daily and ranked by relevance. Distill AI tracks Reinforcement Learning so you don’t have to: get the standout work delivered to your inbox every morning, with 2-sentence summaries and the option to chat with any paper.

Recent papers

• Digital-Twin-Driven Predictive Maintenance and Fault-Tolerant Control for Electrified Agricultural Machinery: A Multiphysics and Deep Reinforcement Learning Framework for PMSM In-Wheel Drives
  Hongyu Xia · PhilPapers (PhilPapers Foun... · Dec 31, 2026

  Electrified agricultural machinery increasingly relies on permanent-magnet synchronous motor drives that must remain dependable under variable traction, terrain-induced vibration, thermal cycling, contamination, and intermittent duty. Exist…

• world2vec: Mechanistic world models through latent dynamics learning
  Tobias Schmidt, Rodrigo Gonzalez Laiz, Krisztián Szuppinger, Steffen Schneider · arXiv · Oct 29, 2026

  World models can achieve strong control performance without faithfully capturing the dynamics of their environment, limiting their potential for mechanistic interpretability and scientific discovery. Here, we introduce world2vec to learn me…

• Training Foundation Models with a Single Algorithm
  Yuda Song, Alvin Zhang, Fahim Tajwar, Daman Arora et al. · arXiv · Oct 29, 2026

  Foundation models are typically trained in stages: pretraining, midtraining, supervised fine-tuning (SFT), and reinforcement learning (RL), with changes in both data and learning algorithms. We show that this algorithmic separation is not f…

• Homebrewed SLMs: Autoregressive and Masked Diffusion Language Models Built at Home for Portuguese
  Pedro Carvalho Brom · arXiv · Oct 29, 2026

  Building a language model of one's own no longer requires a data centre. This tutorial trains one from scratch, tokenizer included, on one consumer GPU, and shows how to tell whether it is good for the tasks of a specialist domain. It propo…

• Reinforcement learning for control of partial differential equations
  Nicolò Botteghi, Matteo Tomasetto, Andrea Manzoni · arXiv · Oct 29, 2026

  Controlling dynamical systems governed by partial differential equations (PDEs) remains a challenging task in scientific computing, owing to high-dimensional state spaces, distributed actuation, nonlinear dynamics, and stringent real-time c…

• Uncertainty-aware Feedback Loop Diagnosis: integrating reinforcement learning and conformal prediction for reliable wafer defect classification
  Aleksandr Dekhovich, Carlas Smith, Oleg Soloviev · arXiv · Oct 29, 2026

  Semiconductor manufacturing is a complex, multi-step process that requires both high precision and low operating costs. Therefore, detecting defects in a high-volume setting is an important component of wafer map analysis for reliable produ…

• Uncertainty Quantification for Markov Chain Induced Martingales with Application to Temporal Difference Learning
  Weichen Wu, Yuting Wei, Alessandro Rinaldo · SIAM Journal on Mathematics... · Oct 28, 2026

  Abstract. We establish novel and general high-dimensional concentration inequalities and Berry–Esseen bounds for vector-valued martingales induced by Markov chains. We apply these results to analyze the performance of the temporal differenc…

• Reinforcement learning and multi-agent systems-driven optimization of emotional perception and collaborative response mechanisms in intelligent customer service systems
  Xuanang Li, Shisong Wu, Yanghao Li, Shengyao Zhuang et al. · Discover Computing · Oct 28, 2026
• EnvSched: Environment-Aware Scheduling for Rollout-Efficient Agent Reinforcement Learning
  Sihan Yang, Weichen Bai, Ruichuan An, Yiming Cheng et al. · arXiv · Oct 27, 2026

  As executable environments for language agents scale, generating and verifying their rollouts becomes increasingly expensive. Under finite rollout budgets, this raises an important question: can data management at the environment level make…

• Collective Dynamics of Self-Reinforcement: How Cross-Task Coupling Drives Policy Collapse in Unsupervised RLVR
  Peize Ding · arXiv · Oct 27, 2026

  Label-free reinforcement learning with verifiable rewards (RLVR) rewards a language model for agreeing with its own majority vote. It improves reasoning at first but eventually collapses: the self-assessed reward goes to one while accuracy …

• Interpretable Meta-Policies for Neural Decision Making
  Raban Emunds, Jannis Blüml, Quentin Delfosse, Kristian Kersting · arXiv · Oct 27, 2026

  Deep Reinforcement Learning remains difficult to interpret and align due to the black-box nature of neural networks. While (neuro-)symbolic approaches improve transparency by incorporating symbolic reasoning, applying them directly to low-l…

• Hierarchical multi-agent reinforcement learning for operations and maintenance disruption recovery in high-end equipment manufacturing under compound risks
  Chen Wei, Ping-feng Liu, Wan-ying Rao · Scientific Reports · Oct 27, 2026
• Drift-Gated Ensemble Pessimism for Offline Meta-Reinforcement Learning Robust to Behavior-Policy Shift
  Chang-Hoon Jeong, Kisung Shin, Byoung-Tak Zhang · Journal of KIISE · Oct 27, 2026
• FOCUS: Active Critic Learning for Long-Horizon Language Model Agents
  Zhengping Jiang, William Jurayj, Liaoyaqi Wang, Chunsheng Zuo et al. · arXiv · Oct 26, 2026

  Sparse outcome rewards make it difficult for long-horizon language-model agents to assign credit to earlier actions. Learned critics provide value estimates for intermediate states, but these estimates can be inaccurate, while estimating in…

• Multi-Objective Reinforcement Learning Using Routed Ensembles and Trajectory Attribution
  Zhenshuo Zhang, Ruoxuan Xiong, Hongyang R. Zhang · arXiv · Oct 26, 2026

  Multi-objective reinforcement learning is a fundamental problem for modern AI, with applications spanning post-training, reasoning models, and robotic control. A central challenge is to learn policies that balance competing objectives, such…

• Audit Where Hacks Pay: Budgeted Strong-Verifier Audits Against Reward Hacking in GRPO
  Shurun Li, Michael Hua Wang, Haibo Zeng, Long Wang · arXiv · Oct 26, 2026

  Reinforcement learning with verifiable rewards trains on cheap checks, such as unit tests, that a policy can learn to game, while a stronger check is too expensive to run on every rollout. We study how to spend a limited strong-verifier bud…

• MiRL: Error-Targeted Reinforcement Learning for Correcting Terminal Agent Misbehaviors
  Runchu Tian, Zaid Khan, Zun Wang, Justin Chih-Yao Chen et al. · arXiv · Oct 26, 2026

  Error-targeted learning gives a legible way to improve a model: identify specific weaknesses of the model, train the model to fix these weaknesses, and check whether the weaknesses are fixed. We study how to apply error-targeted learning to…

• Provably Efficient Q-learning for Robust Reinforcement Learning
  Yuchen Jiao, Gen Li · arXiv · Oct 26, 2026

  Standard reinforcement learning (RL) algorithms based on Markov Decision Processes (MDPs) are vulnerable to sim-to-real gaps. To address this issue, distributionally robust RL algorithms based on robust MDPs (RMDPs) have been proposed, for …

• TLDQ: time-layer joint-aware reinforcement learning dynamic quantization algorithm for seismic phase picking on resource-constrained edge devices
  Jianxian Cai, Jiapeng Zhou, Yan Shi, Dongyan Cui · Earth Science Informatics · Oct 26, 2026
• Reinforcement-learning irrigation scheduling under soil-moisture sensor dropout: a climate-analytics evaluation of yield-regret bounds and groundwater-drawdown equity
  Isatou Jallow-Njie, Ebrima Sanyang · Zenodo (CERN European Organ... · Oct 24, 2026

  Soil-moisture sensor dropout is a routine operational failure in smallholder irrigation schemes, yet reinforcement-learning (RL) schedulers are typically evaluated under the assumption of continuous state observation. The framework yields t…

• Counterfactual sensitivity analysis of reinforcement-learning irrigation schedulers under unobserved soil-moisture states: a climate-analytics evaluation of water-allocation regret
  Khadidiatou Sow · Zenodo (CERN European Organ... · Oct 24, 2026

  Reinforcement-learning (RL) irrigation schedulers are trained and evaluated against soil-moisture trajectories that, in field deployment, are only partially observed. The framework is methodological and theoretical; it does not report a fie…

• Perceived Implementation of Reinforcement Strategies and Academic Performance of Intermediate Pupils in Lutayan District II
  Jurine Mae Legarde, Mikee Pancho, Nathaniel D. Naanep, Jonathan Roque · International Journal of Tr... · Oct 24, 2026

  This study examined the implementation of reinforcement strategies (verbal praise, rewards and incentives, and corrective feedback) as rated by pupils, teachers, and school heads, and the relationship between pupils’ perceived exposure to t…

• Multi-agent reinforcement learning for coordinated reservoir releases under conflicting hydropower and irrigation mandates: a climate-analytics evaluation of Pareto-regret and downstream equity
  Aïchatou Djibo-Boubacar, Mariama Hama Abdou, Issoufou Soumana Oumarou · Zenodo (CERN European Organ... · Oct 24, 2026

  Reservoir operators in transboundary basins face a structural conflict: hydropower revenue is maximised by releasing water when electricity prices peak, while irrigation mandates require deliveries timed to crop water demand. Multi-agent re…

• Multi-agent reinforcement learning for reservoir cascade release scheduling: a climate-analytics evaluation of downstream low-flow equity under forecast uncertainty
  Hiwot Tesfaye Wolde, Selamawit Gebremedhin Desta, Mulugeta Assefa · Zenodo (CERN European Organ... · Oct 24, 2026

  Cascade reservoir operators must release water under uncertain inflow forecasts while satisfying competing demands for hydropower, irrigation and downstream environmental flows. The framework is formalised as a constrained Markov game with …

• RobustRLlib: A Unified Library and Benchmark for Robust Reinforcement Learning Algorithms
  Xi Chen, Fengqi Hu, Zeyu Shen, Jiaying Xu et al. · arXiv · Oct 23, 2026

  Robust reinforcement learning (RL) is critical for reliable real-world deployment, where policies may fail under diverse shifts and disruptions. Existing robust RL algorithms offer a rich set of mechanisms and principles, yet two major barr…

• Where Rounding Happens Matters: Rounding-Location Alignment Stabilizes LLM Reinforcement Learning
  Yijing Ke, Tianle Cai, Ruosong Wang · arXiv · Oct 23, 2026

  Reinforcement learning for large language models often uses separate systems for inference and training. Even with identical weights and BF16 tensors, implementation differences can make an intended on-policy update off-policy and destabili…

• Recommender regret for improving regularised RL estimation with bandits and MCTS
  Danilo Vucetic, Alexandre Larouche, Mohammad Pedramfar, Audrey Durand et al. · arXiv · Oct 23, 2026

  Regularised reinforcement learning (RL) algorithms rely on estimators, like Monte Carlo tree search or one-step look-ahead, to provide policy, value, or Q-value estimates in training and inference. However, existing bandit-based estimation …

• Reinforcement Learning with Multi-Step Transition Predictions: A Bellman Theory
  Chenbei Lu, Zaiwei Chen, Tongxin Li, Chenye Wu et al. · arXiv · Oct 23, 2026

  Multi-step predictions of future transition realizations can improve reinforcement learning, but their value remains difficult to quantify. Directly including these predictions in the state can lead to exponential state-space growth, making…

• Regularised RL improves sim-to-real transfer
  Danilo Vucetic, Mohamad H. Danesh, Florence Clerc, Hsiu-Chin Lin et al. · arXiv · Oct 23, 2026

  Regularised reinforcement learning (RL) has provable robustness guarantees that should ease the transfer of policies from simulation to robots, known as sim-to-real. Yet, weakly regularised algorithms like Proximal Policy Optimisation (PPO)…

• METTA: Memory-Enhanced Test-Time Adaptation for Meta-Reinforcement Learning
  Ali Najar, Hosein Hasani, Amir Kooshan Fattah, Yasamin Rajabi et al. · arXiv · Oct 22, 2026

  Meta-reinforcement learning aims to train agents that adapt quickly to unseen tasks from limited interaction, but this requires representing task evidence over long interaction histories. Classical recurrent agents compress experience into …