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 379 篇，第 7 / 8 页

301. ICLR 2025Robust Transfer of Safety-Constrained Reinforcement Learning Agents
302. ICLR 2025RTDiff: Reverse Trajectory Synthesis via Diffusion for Offline Reinforcement Learning
303. ICLR 2025Safety Representations for Safer Policy Learning
304. ICLR 2025Safety-Prioritizing Curricula for Constrained Reinforcement Learning
305. ICLR 2025Sail into the Headwind: Alignment via Robust Rewards and Dynamic Labels against Reward Hacking
306. ICLR 2025Scalable Decision-Making in Stochastic Environments through Learned Temporal Abstraction
307. ICLR 2025Scaling Offline Model-Based RL via Jointly-Optimized World-Action Model Pretraining
308. ICLR 2025Scrutinize What We Ignore: Reining In Task Representation Shift Of Context-Based Offline Meta Reinforcement Learning
309. ICLR 2025Seeing Eye to AI: Human Alignment via Gaze-Based Response Rewards for Large Language Models
310. ICLR 2025Select before Act: Spatially Decoupled Action Repetition for Continuous Control
311. ICLR 2025SELF-EVOLVED REWARD LEARNING FOR LLMS
312. ICLR 2025Self-Improving Robust Preference Optimization
313. ICLR 2025Self-Play Preference Optimization for Language Model Alignment
314. ICLR 2025Self-play with Execution Feedback: Improving Instruction-following Capabilities of Large Language Models
315. ICLR 2025Semantic Temporal Abstraction via Vision-Language Model Guidance for Efficient Reinforcement Learning
316. ICLR 2025SEMDICE: Off-policy State Entropy Maximization via Stationary Distribution Correction Estimation
317. ICLR 2025Sequential Stochastic Combinatorial Optimization Using Hierarchal Reinforcement Learning
318. ICLR 2025SeRA: Self-Reviewing and Alignment of LLMs using Implicit Reward Margins
319. ICLR 2025SFESS: Score Function Estimators for $k$-Subset Sampling
320. ICLR 2025SFS: Smarter Code Space Search improves LLM Inference Scaling
321. ICLR 2025Sharpness-Aware Black-Box Optimization
322. ICLR 2025SimBa: Simplicity Bias for Scaling Up Parameters in Deep Reinforcement Learning
323. ICLR 2025Simple, Good, Fast: Self-Supervised World Models Free of Baggage
324. ICLR 2025Simplifying Deep Temporal Difference Learning
325. ICLR 2025SmartRAG: Jointly Learn RAG-Related Tasks From the Environment Feedback
326. ICLR 2025Solving hidden monotone variational inequalities with surrogate losses
327. ICLR 2025SOO-Bench: Benchmarks for Evaluating the Stability of Offline Black-Box Optimization
328. ICLR 2025Sparse Autoencoders Reveal Temporal Difference Learning in Large Language Models
329. ICLR 2025SRSA: Skill Retrieval and Adaptation for Robotic Assembly Tasks
330. ICLR 2025Stabilizing Reinforcement Learning in Differentiable Multiphysics Simulation
331. ICLR 2025Stable Hadamard Memory: Revitalizing Memory-Augmented Agents for Reinforcement Learning
332. ICLR 2025Statistical Tractability of Off-policy Evaluation of History-dependent Policies in POMDPs
333. ICLR 2025Strategist: Self-improvement of LLM Decision Making via Bi-Level Tree Search
334. ICLR 2025Student-Informed Teacher Training
335. ICLR 2025Studying the Interplay Between the Actor and Critic Representations in Reinforcement Learning
336. ICLR 2025Subtask-Aware Visual Reward Learning from Segmented Demonstrations
337. ICLR 2025Swift Hydra: Self-Reinforcing Generative Framework for Anomaly Detection with Multiple Mamba Models
338. ICLR 2025Synthesizing Programmatic Reinforcement Learning Policies with Large Language Model Guided Search
339. ICLR 2025Tackling Data Corruption in Offline Reinforcement Learning via Sequence Modeling
340. ICLR 2025Taming Overconfidence in LLMs: Reward Calibration in RLHF
341. ICLR 2025Temporal Difference Learning: Why It Can Be Fast and How It Will Be Faster
342. ICLR 2025The "Law'' of the Unconscious Contrastive Learner: Probabilistic Alignment of Unpaired Modalities
343. ICLR 2025The Value of Sensory Information to a Robot
344. ICLR 2025ToolGen: Unified Tool Retrieval and Calling via Generation
345. ICLR 2025TOP-ERL: Transformer-based Off-Policy Episodic Reinforcement Learning
346. ICLR 2025Toward Efficient Multi-Agent Exploration With Trajectory Entropy Maximization
347. ICLR 2025Toward Exploratory Inverse Constraint Inference with Generative Diffusion Verifiers
348. ICLR 2025Towards Empowerment Gain through Causal Structure Learning in Model-Based Reinforcement Learning
349. ICLR 2025Towards Federated RLHF with Aggregated Client Preference for LLMs
350. ICLR 2025Towards General-Purpose Model-Free Reinforcement Learning

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。