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 12 / 138 页

551. ICLR 2026FAPO: Flawed-Aware Policy Optimization for Efficient and Reliable Reasoning
552. ICLR 2026FastGRPO: Accelerating Policy Optimization via Concurrency-aware Speculative Decoding and Online Draft Learning
553. ICLR 2026Fathom-DeepResearch: Unlocking Long Horizon Information Retrieval and Synthesis for SLMs
554. ICLR 2026Federated ADMM from Bayesian Duality
555. ICLR 2026Fine-R1: Make Multi-modal LLMs Excel in Fine-Grained Visual Recognition by Chain-of-Thought Reasoning
556. ICLR 2026Fine-tuning Behavioral Cloning Policies with Preference‑Based Reinforcement Learning
557. ICLR 2026Fine-Tuning Diffusion Models via Intermediate Distribution Shaping
558. ICLR 2026Finite-Time Analysis of Actor-Critic Methods with Deep Neural Network Approximation
559. ICLR 2026FIRE: Frobenius-Isometry Reinitialization for Balancing the Stability–Plasticity Tradeoff
560. ICLR 2026floq: Training Critics via Flow-Matching for Scaling Compute in Value-Based RL
561. ICLR 2026Flow Actor-Critic for Offline Reinforcement Learning
562. ICLR 2026Flow Matching Policy Gradients
563. ICLR 2026Flow Matching with Injected Noise for Offline-to-Online Reinforcement Learning
564. ICLR 2026Flow-Based Single-Step Completion for Efficient and Expressive Policy Learning
565. ICLR 2026Flowing Through States: Neural ODE Regularization for Reinforcement Learning
566. ICLR 2026FlowRL: Matching Reward Distributions for LLM Reasoning
567. ICLR 2026Foundational Automatic Evaluators: Scaling Multi-Task Generative Evaluator Training for Reasoning-Centric Domains
568. ICLR 2026FrameThinker: Learning to Think with Long Videos via Multi-Turn Frame Spotlighting
569. ICLR 2026From Broad Exploration to Stable Synthesis: Entropy-Guided Optimization for Autoregressive Image Generation
570. ICLR 2026From Curiosity to Caution: Mitigating Reward Hacking for Best-of-$N$ with Pessimism
571. ICLR 2026From f(x) and g(x) to f(g(x)): LLMs Learn New Skills in RL by Composing Old Ones
572. ICLR 2026From Gradient Volume to Shapley Fairness: Towards Fair Multi-Task Learning
573. ICLR 2026From Language to Locomotion: Retargeting-free Humanoid Control via Motion Latent Guidance
574. ICLR 2026From Narrow to Panoramic Vision: Attention-Guided Cold-Start Reshapes Multimodal Reasoning
575. ICLR 2026From Observations to Events: Event-Aware World Models for Reinforcement Learning
576. ICLR 2026From Parameters to Behaviors: Unsupervised Compression of the Policy Space
577. ICLR 2026From Seeing to Experiencing: Scaling Navigation Foundation Models with Reinforcement Learning
578. ICLR 2026From Sequential to Parallel: Reformulating Dynamic Programming as GPU Kernels for Large-Scale Stochastic Combinatorial Optimization
579. ICLR 2026From Sorting Algorithms to Scalable Kernels: Bayesian Optimization in High-Dimensional Permutation Spaces
580. ICLR 2026From Ticks to Flows: Dynamics of Neural Reinforcement Learning in Continuous Environments
581. ICLR 2026From Verifiable Dot to Reward Chain: Harnessing Verifiable Reference-based Rewards for Reinforcement Learning of Open-ended Generation
582. ICLR 2026Frozen Policy Iteration: Computationally Efficient RL under Linear $Q^{\pi}$ Realizability for Deterministic Dynamics
583. ICLR 2026FrugalRAG: Less is More in RL Finetuning for Multi-hop Question Answering
584. ICLR 2026Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments
585. ICLR 2026GALAX: Graph-Augmented Language Model for Explainable Reinforcement-Guided Subgraph Reasoning in Precision Medicine
586. ICLR 2026GAR: Generative Adversarial Reinforcement Learning for Formal Theorem Proving
587. ICLR 2026GAS: Enhancing Reward-Cost Balance of Generative Model-assisted Offline Safe RL
588. ICLR 2026GDPval: Evaluating AI Model Performance on Real-World Economically Valuable Tasks
589. ICLR 2026GEM: A Gym for Generalist LLMs
590. ICLR 2026General Exploratory Bonus for Optimistic Exploration in RLHF
591. ICLR 2026Generalizable End-to-End Tool-Use RL with Synthetic CodeGym
592. ICLR 2026Generalization of RLVR Using Causal Reasoning as a Testbed
593. ICLR 2026Generalized Parallel Scaling with Interdependent Generations
594. ICLR 2026Generative Adversarial Post-Training Mitigates Reward Hacking in Live Human-AI Music Interaction
595. ICLR 2026Generative Adversarial Reasoner: Enhancing LLM Reasoning with Adversarial Reinforcement Learning
596. ICLR 2026Geometric-Mean Policy Optimization
597. ICLR 2026Geometry of Uncertainty: Learning Metric Spaces for Multimodal State Estimation in RL
598. ICLR 2026GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning
599. ICLR 2026GEPO: Group Expectation Policy Optimization for Stable Heterogeneous Reinforcement Learning
600. ICLR 2026Getting Your LLMs Ready for Reinforcement Learning with Lightweight SFT

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。