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 2,494 篇，第 6 / 50 页

251. NeurIPS 2026From POMDP Theory to Deep RL with Particle Filters
252. NeurIPS 2026From Reasoning Chains to Verifiable Subproblems: Curriculum Reinforcement Learning Enables Credit Assignment for LLM Reasoning
253. NeurIPS 2026From Sample to Subset Construction: Coverage-Aware Curation of Robot Demonstrations
254. NeurIPS 2026From Structural Feedback to Prompt Policies: Learning Faithful Text-to-Image Prompt Editors
255. NeurIPS 2026From Views to Worlds: Active Exploration over 3D Worlds for Vision-Language Models
256. NeurIPS 2026FuRA: Full-Rank Parameter-Efficient Fine-Tuning with Spectral Preconditioning
257. NeurIPS 2026GARDO: Reinforcing Diffusion Models without Reward Hacking
258. NeurIPS 2026GCD: Correcting Hidden-State Bias in Off-Policy Agentic RL
259. NeurIPS 2026GDMD: Guiding Distribution Matching Distillation with Gradient-Based Reinforcement Learning
260. NeurIPS 2026GDSD: Reinforcement Learning as Guided Denoiser Self-Distillation for Diffusion Language Models
261. NeurIPS 2026Gen-Searcher: Reinforcing Agentic Search for Image Generation
262. NeurIPS 2026Generalised Linear Models in Deep Bayesian RL with Learnable Basis Functions
263. NeurIPS 2026Generalized Intention Modeling in Multi-Agent Reinforcement Learning
264. NeurIPS 2026Generation Navigator: A State-Aware Agentic Framework for Image Generation
265. NeurIPS 2026Generative Actor-Critic with Soft Bridge Policies
266. NeurIPS 2026Generative Scenario Rollouts for End-to-End Autonomous Driving
267. NeurIPS 2026GeoSym127K: Scalable Symbolically-verifiable Synthesis for Multimodal Geometric Reasoning
268. NeurIPS 2026GeoX: Mastering Geospatial Reasoning Through Self-Play and Verifiable Rewards
269. NeurIPS 2026GNES: Neural-Guided Evolutionary Program Search for Interpretable Multi-Agent Control
270. NeurIPS 2026Gradient Routing Localizes and Removes Unintended Behaviors in RL
271. NeurIPS 2026Graph Topology Augmentation for Prioritized Sweeping in Non-stationary Reinforcement Learning
272. NeurIPS 2026GraphMemRL: Action-Native Reinforcement Learning for Persistent Graph Memory Construction
273. NeurIPS 2026Grounding Agent Reasoning with Structured Process Supervision for Multi-turn Reinforcement Learning
274. NeurIPS 2026Grounding Multimodal Reasoning with Evidence-Ablated Negatives
275. NeurIPS 2026Group-Graph Policy Optimization for Long-Horizon Agentic Reinforcement Learning
276. NeurIPS 2026Grouped Adaptive Head Mixing for Personalized Multi-Task Federated Reinforcement Learning
277. NeurIPS 2026Hadamard Representation: Scaffolding Performance Across Model-free RL
278. NeurIPS 2026Hearing the Unspoken: Simulator-Induced Asymmetric-View Policy Optimization for Proactive Task-Oriented Dialogue
279. NeurIPS 2026Heterogeneous Agent Collaborative Reinforcement Learning
280. NeurIPS 2026Hide to Guide: Learning via Semantic Masking
281. NeurIPS 2026Hider–Seeker Self-Play: Geometry-Verifiable Process Rewards for Long-Horizon Visual Search
282. NeurIPS 2026Hierarchical Concept Geometry in Language Representations Emerges from Word Co-occurrence
283. NeurIPS 2026HiFloat4 Format for End-To-End Reinforcement Learning Post-Training of Large Language Models
284. NeurIPS 2026High-Dimensional Robotic Reinforcement Learning with Developing Synergies
285. NeurIPS 2026Higher-Order Action Supervision Makes A Strong Policy Class
286. NeurIPS 2026HiPhy: Hierarchical Alignment for Physically-Plausible Multi-Principle Video Generation
287. NeurIPS 2026Hitting Time Isomorphism for Multi-Stage Planning with Foundation Policies
288. NeurIPS 2026Hodge Laplacian Quasi-Harmonic Flows for Option Discovery
289. NeurIPS 2026HopChain: Multi-Hop Data Synthesis for Generalizable Vision-Language Reasoning
290. NeurIPS 2026HOPE: Hand-Object Pressure Estimation from Monocular Videos
291. NeurIPS 2026Horizon Adaptive Offline Policy Learning via Value Stitching
292. NeurIPS 2026How does RL Post-training Induce Skill Composition? A Case Study on Countdown
293. NeurIPS 2026How Neural Reward Models Learn Features for Policy Optimization: A Single-Index Analysis
294. NeurIPS 2026Hybrid-LoRA: Bridging Full Fine-Tuning and Low-Rank Adaptation for Post-Training
295. NeurIPS 2026IFDECORATOR: Wrapping Instruction Following Reinforcement Learning with Verifiable Rewards
296. NeurIPS 2026Imitation Dominates Reinforcement: Direct In-Context RL Is Closer to ICL Than RL
297. NeurIPS 2026Imperfect World Models are Exploitable
298. NeurIPS 2026Improved Regret Analysis For Parallel Gaussian Process Bandit Optimization
299. NeurIPS 2026Improved Regret bounds in Tabular Reinforcement Learning under Local Differential Privacy
300. NeurIPS 2026Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization

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。