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 6,885 篇，第 15 / 138 页

701. ICLR 2026Learning More with Less: A Dynamic Dual-Level Down-Sampling Framework for Efficient Policy Optimization
702. ICLR 2026Learning Nonlinear Causal Reductions to Explain Reinforcement Learning Policies
703. ICLR 2026Learning To Draft: Adaptive Speculative Decoding with Reinforcement Learning
704. ICLR 2026Learning to Generate Unit Test via Adversarial Reinforcement Learning
705. ICLR 2026Learning to Lie: Adversarial Attacks on Human-AI Teams and LLMs
706. ICLR 2026Learning to Orchestrate Agents in Natural Language with the Conductor
707. ICLR 2026Learning to Reason as Action Abstractions with Scalable Mid-Training RL
708. ICLR 2026Learning to Reason Efficiently with Discounted Reinforcement Learning
709. ICLR 2026Learning to Reason for Hallucination Span Detection
710. ICLR 2026Learning to Reason in Structured In-context Environments with Reinforcement Learning
711. ICLR 2026Learning to Reason over Continuous Tokens with Reinforcement Learning
712. ICLR 2026Learning to Reason without External Rewards
713. ICLR 2026Learning to summarize user information for personalized reinforcement learning from human feedback
714. ICLR 2026Learning What Reinforcement Learning Can't: Interleaved Online Fine-Tuning for Hardest Questions
715. ICLR 2026Less Is More: Clustered Cross-Covariance Control for Offline RL
716. ICLR 2026Leveraging Discrete Function Decomposability for Scientific Design
717. ICLR 2026Leveraging Explanation to Improve Generalization of Meta Reinforcement Learning
718. ICLR 2026Linking Process to Outcome: Conditional Reward Modeling for LLM Reasoning
719. ICLR 2026LLMs are Greedy Agents: Effects of RL Fine-tuning on Decision-Making Abilities
720. ICLR 2026LLMs Must Think Thrice to Solve Executable Counterfactuals
721. ICLR 2026Local Reinforcement Learning with Action-Conditioned Root Mean Squared Q-Functions
722. ICLR 2026Long-Document QA with Chain-of-Structured-Thought and Fine-Tuned SLMs
723. ICLR 2026LongRLVR: Long-Context Reinforcement Learning Requires Verifiable Context Rewards
724. ICLR 2026LongWriter-Zero: Mastering Ultra-Long Text Generation via Reinforcement Learning
725. ICLR 2026Look Back to Reason Forward: Revisitable Memory for Long-Context LLM Agents
726. ICLR 2026Look-ahead Reasoning with a Learned Model in Imperfect Information Games
727. ICLR 2026Lookahead Tree-Based Rollouts for Enhanced Trajectory-Level Exploration in Reinforcement Learning with Verifiable Rewards
728. ICLR 2026LoongRL: Reinforcement Learning for Advanced Reasoning over Long Contexts
729. ICLR 2026MAC-AMP: A Closed-Loop Multi-Agent Collaboration System for Multi-Objective Antimicrobial Peptide Design
730. ICLR 2026MAGE: Multi-scale Autoregressive Generation for Offline Reinforcement Learning
731. ICLR 2026MAGO: Beyond Fixed Hyperparameters with Multi-Objective Pareto Optimization for Hybrid LLM Reasoning
732. ICLR 2026Making Slow Thinking Faster: Compressing LLM Chain-of-Thought via Step Entropy
733. ICLR 2026MARC: Memory-Augmented RL Token Compression for Efficient Video Understanding
734. ICLR 2026Markovian Transformers for Informative Language Modeling
735. ICLR 2026MARS-Sep: Multimodal-Aligned Reinforced Sound Separation
736. ICLR 2026MARSHAL: Incentivizing Multi-Agent Reasoning via Self-Play with Strategic LLMs
737. ICLR 2026MARTI: A Framework for Multi-Agent LLM Systems Reinforced Training and Inference
738. ICLR 2026Masked Skill Token Training for Hierarchical Off-Dynamics Transfer
739. ICLR 2026Master Skill Learning with Policy-Grounded Synergy of LLM-based Reward Shaping and Exploring
740. ICLR 2026Mastering Sparse CUDA Generation through Pretrained Models and Deep Reinforcement Learning
741. ICLR 2026MATH-Beyond: A Benchmark for RL to Expand Beyond the Base Model
742. ICLR 2026Mathesis: Towards Formal Theorem Proving from Natural Languages
743. ICLR 2026Mean Flow Policy with Instantaneous Velocity Constraint for One-step Action Generation
744. ICLR 2026Measuring and Mitigating Rapport Bias of Large Language Models under Multi-Agent Social Interactions
745. ICLR 2026Measuring Audio's Impact on Correctness: Audio-Contribution-Aware Post-Training of Large Audio Language Models
746. ICLR 2026Mechanistic Detection and Mitigation of Hallucination in Large Reasoning Models
747. ICLR 2026MedAgentGym: A Scalable Agentic Training Environment for Code-Centric Reasoning in Biomedical Data Science
748. ICLR 2026MedVR: Annotation-Free Medical Visual Reasoning via Agentic Reinforcement Learning
749. ICLR 2026MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents
750. ICLR 2026Memory-T1: Reinforcement Learning for Temporal Reasoning in Multi-session Agents

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。