强化学习论文趋势 2023–2026

主要结论

• 强化学习论文从 2023 年的 1,070 篇增长到 2026 年的 2,985 篇，三年间扩大到约 2.8 倍。
• NeurIPS 是发强化学习论文最多的会议，2023–2026 年共 2,494 篇，占四个会议总量的 36%。
• 交叉方向里涨得最快的是推理模型：同时属于两个主题的论文从 2023 年的 9 篇增至 2026 年的 437 篇。

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

• NeurIPS 2026\texttt{FEROM}: Frontier Endogenous Reveal-Order Marginal Policy Optimization for Masked Diffusion LMs
• ICML 2026$\text{DT}^\text{2}$: Decision-Targeted Digital Twins
• ICLR 2026$\nabla$-Reasoner: LLM Reasoning via Test-Time Gradient Descent in Textual Space
• CVPR 2026A Causal Marriage between VLM and IRM from Understanding to
• NeurIPS 2025$\texttt{G1}$: Teaching LLMs to Reason on Graphs with Reinforcement Learning
• ICML 2025A Computationally Efficient Algorithm for Infinite-Horizon Average-Reward Linear MDPs
• ICLR 2025$\phi$-Update: A Class of Policy Update Methods with Policy Convergence Guarantee
• CVPR 2025All-Optical Nonlinear Diffractive Deep Network for Ultrafast Image Denoising
• NeurIPS 2024A Critical Evaluation of AI Feedback for Aligning Large Language Models
• ICML 2024${\rm E}(3)$-Equivariant Actor-Critic Methods for Cooperative Multi-Agent Reinforcement Learning
• ICLR 2024A Black-box Approach for Non-stationary Multi-agent Reinforcement Learning
• CVPR 2024DMR: Decomposed Multi-Modality Representations for Frames and Events Fusion in Visual Reinforcement Learning
• NeurIPS 2023“Why Not Looking backward?” A Robust Two-Step Method to Automatically Terminate Bayesian Optimization
• ICML 2023A Connection between One-Step RL and Critic Regularization in Reinforcement Learning
• ICLR 2023$O(T^{-1})$ Convergence of Optimistic-Follow-the-Regularized-Leader in Two-Player Zero-Sum Markov Games
• CVPR 2023"Seeing" Electric Network Frequency From Events
• NeurIPS 2026$\varphi$TD: Distributional Reinforcement Learning using Characteristic Functions
• ICML 2026$\textit{S}$-SPPO: Semantic-Calibrated Self-Play Preference Optimization
• ICLR 2026$AutoDrive\text{-}P^3$: Unified Chain of Perception–Prediction–Planning Thought via Reinforcement Fine-Tuning
• CVPR 2026AdaptVision Efficient Vision-Language Models via Adaptive Visual Acquisition

全部强化学习论文

共 1,827 篇，第 19 / 37 页

901. ICLR 2026When Silence Is Golden: Can LLMs Learn to Abstain in Temporal QA and Beyond?
902. ICLR 2026Who Matters Matters: Agent-Specific Conservative Offline MARL
903. ICLR 2026WholeBodyVLA: Towards Unified Latent VLA for Whole-body Loco-manipulation Control
904. ICLR 2026Why DPO is a Misspecified Estimator and How to Fix It
905. ICLR 2026Wiki-R1: Incentivizing Multimodal Reasoning for Knowledge-based VQA via Data and Sampling Curriculum
906. ICLR 2026WIMLE: Uncertainty‑Aware World Models with IMLE for Sample‑Efficient Continuous Control
907. ICLR 2026WMPO: World Model-based Policy Optimization for Vision-Language-Action Models
908. ICLR 2026XQC: Well-conditioned Optimization Accelerates Deep Reinforcement Learning
909. ICLR 2026Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking
910. ICLR 2026Zebra-CoT: A Dataset for Interleaved Vision-Language Reasoning
911. ICLR 2026Zero-Shot Adaptation of Behavioral Foundation Models to Unseen Dynamics
912. ICLR 2025$\phi$-Update: A Class of Policy Update Methods with Policy Convergence Guarantee
913. ICLR 2025$q$-exponential family for policy optimization
914. ICLR 20253DMolFormer: A Dual-channel Framework for Structure-based Drug Discovery
915. ICLR 2025A Black Swan Hypothesis: The Role of Human Irrationality in AI Safety
916. ICLR 2025A Causal Lens for Learning Long-term Fair Policies
917. ICLR 2025A Common Pitfall of Margin-based Language Model Alignment: Gradient Entanglement
918. ICLR 2025A Distributional Approach to Uncertainty-Aware Preference Alignment Using Offline Demonstrations
919. ICLR 2025A Generalist Hanabi Agent
920. ICLR 2025A Policy-Gradient Approach to Solving Imperfect-Information Games with Best-Iterate Convergence
921. ICLR 2025A Single Goal is All You Need: Skills and Exploration Emerge from Contrastive RL without Rewards, Demonstrations, or Subgoals
922. ICLR 2025A Theoretical Framework for Partially-Observed Reward States in RLHF
923. ICLR 2025Accelerating Goal-Conditioned Reinforcement Learning Algorithms and Research
924. ICLR 2025Accelerating Task Generalisation with Multi-Level Skill Hierarchies
925. ICLR 2025Accelerating Training with Neuron Interaction and Nowcasting Networks
926. ICLR 2025Action abstractions for amortized sampling
927. ICLR 2025Actions Speak Louder Than Words: Rate-Reward Trade-off in Markov Decision Processes
928. ICLR 2025ACTIVE: Offline Reinforcement Learning via Adaptive Imitation and In-sample $V$-Ensemble
929. ICLR 2025ActSafe: Active Exploration with Safety Constraints for Reinforcement Learning
930. ICLR 2025Ada-K Routing: Boosting the Efficiency of MoE-based LLMs
931. ICLR 2025Adaptive $Q$-Network: On-the-fly Target Selection for Deep Reinforcement Learning
932. ICLR 2025Adaptive Gradient Clipping for Robust Federated Learning
933. ICLR 2025Adaptive teachers for amortized samplers
934. ICLR 2025AdaWM: Adaptive World Model based Planning for Autonomous Driving
935. ICLR 2025Adding Conditional Control to Diffusion Models with Reinforcement Learning
936. ICLR 2025Advantage Alignment Algorithms
937. ICLR 2025Adversarial Generative Flow Network for Solving Vehicle Routing Problems
938. ICLR 2025Adversarial Policy Optimization for Offline Preference-based Reinforcement Learning
939. ICLR 2025AHA: A Vision-Language-Model for Detecting and Reasoning Over Failures in Robotic Manipulation
940. ICLR 2025Aligning Visual Contrastive learning models via Preference Optimization
941. ICLR 2025An Evolved Universal Transformer Memory
942. ICLR 2025An Optimal Discriminator Weighted Imitation Perspective for Reinforcement Learning
943. ICLR 2025Any-step Dynamics Model Improves Future Predictions for Online and Offline Reinforcement Learning
944. ICLR 2025As Simple as Fine-tuning: LLM Alignment via Bidirectional Negative Feedback Loss
945. ICLR 2025Asynchronous Federated Reinforcement Learning with Policy Gradient Updates: Algorithm Design and Convergence Analysis
946. ICLR 2025Asynchronous RLHF: Faster and More Efficient Off-Policy RL for Language Models
947. ICLR 2025BALROG: Benchmarking Agentic LLM and VLM Reasoning On Games
948. ICLR 2025BAMDP Shaping: a Unified Framework for Intrinsic Motivation and Reward Shaping
949. ICLR 2025Behavioral Entropy-Guided Dataset Generation for Offline Reinforcement Learning
950. ICLR 2025Beyond-Expert Performance with Limited Demonstrations: Efficient Imitation Learning with Double Exploration

常见问题

2023–2026 年四大会议一共有多少篇强化学习论文？

共 6,885 篇，按年份分别是 2023 年 1,070 篇、2024 年 1,190 篇、2025 年 1,640 篇、2026 年 2,985 篇。

NeurIPS 2026 有多少篇强化学习论文？

NeurIPS 2026 有 760 篇强化学习论文，上一届是 784 篇。

ICLR 2026 有多少篇强化学习论文？

ICLR 2026 有 911 篇强化学习论文，上一届是 379 篇。

ICML 2026 有多少篇强化学习论文？

ICML 2026 有 1,021 篇强化学习论文，上一届是 405 篇。

CVPR 2026 有多少篇强化学习论文？

CVPR 2026 有 293 篇强化学习论文，上一届是 72 篇。

哪个会议的强化学习论文最多？

NeurIPS，2023–2026 年共 2,494 篇。