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 3 / 50 页

101. NeurIPS 2026CLIFT: Conformal Self-Verification for Web Agent Training and Test-Time Scaling
102. NeurIPS 2026Co-Evolving Policy Distillation
103. NeurIPS 2026Co-GRPO: Co-Optimized Group Relative Policy Optimization for Masked Image Generation
104. NeurIPS 2026Code2World: A GUI World Model via Renderable Code Generation
105. NeurIPS 2026CodeScaler: Scaling Code LLM Training and Test-Time Inference via Reward Models
106. NeurIPS 2026CoE-Agent: Co-Evolving Patient-Doctor Agents via Interactive Policy Graph Optimization for Clinical Decision Making
107. NeurIPS 2026Cognitive Firewalls: A Synthetic Account of Cross-Lingual Reasoning Collapse
108. NeurIPS 2026COMPASS: Composable Policy-Amortized Structured Search for LLM-Based Optimization Modeling
109. NeurIPS 2026CompilerKV: Risk-Adaptive KV Cache Compression via Offline Experience Compilation
110. NeurIPS 2026Compose Your Oracles: Off Policy Improvement with Aggregated Guidance
111. NeurIPS 2026Composition-RL: Compose Your Verifiable Prompts for Reinforcement Learning of Large Language Models
112. NeurIPS 2026Compute Aligned Training: Optimizing for Test Time Inference
113. NeurIPS 2026Conclusions from Circuit Extraction Depend on the Level of Description: A Controlled Comparison
114. NeurIPS 2026Conformal Selective Acting: Anytime-Valid Risk Control for RLVR-Trained LLMs
115. NeurIPS 2026Conservatism Controllable Compositional Guidance for Offline Safe Reinforcement Learning
116. NeurIPS 2026Conservative Pareto Set Amortization for Offline Multi-Objective Optimization
117. NeurIPS 2026Constrained Goal-directed Planar Graph Generation with Grammar-based Reinforcement Learning
118. NeurIPS 2026Context Value Informed In-Context Reinforcement Learning
119. NeurIPS 2026Control First, Robustness Next: Decoupled Representation Learning for Visual RL Generalization
120. NeurIPS 2026Controllable Molecular Generative Foundation Models
121. NeurIPS 2026Controllable Multi-label Video Safety Detection via Adaptive Tversky Policy Optimization
122. NeurIPS 2026Cooperative Multi-Agent Reinforcement Learning via Epigraph-Form Guided Exploration
123. NeurIPS 2026Coordinating Hundreds of RL Agents through Scalable Inference-Time Search
124. NeurIPS 2026Coordination Connectivity: Shared Initialization Shapes the Joint-Policy Landscape in MARL
125. NeurIPS 2026Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation
126. NeurIPS 2026Correct, Route, Calibrate: Efficient Preference Optimization from Noisy, Heterogeneous Human Feedback
127. NeurIPS 2026Correctable Fork Tokens: Verifier-Anchored Selective Credit Assignment for Tool-Integrated RLVR
128. NeurIPS 2026Cost-Aware Learning
129. NeurIPS 2026CoT-Guard: Small Models for Strong Monitoring
130. NeurIPS 2026Counterfactual Distillation: Internalizing Reflective Experience into LLM Agent
131. NeurIPS 2026Counterfactual Rollout Replay: Forkable Environments as Free Process Rewards for Software Engineering Agents
132. NeurIPS 2026Coupling-Aware Reinforcement Learning for Co-Evolving Graph Games
133. NeurIPS 2026CPA: Efficient and Stable FP4 RL Training via Cross-Precision Alignment
134. NeurIPS 2026CRAFT: Counterfactual-to-Interactive Reinforcement Fine-Tuning for Driving Policies
135. NeurIPS 2026Credit Assignment with Resets in Language Model Reasoning
136. NeurIPS 2026Cross-Question Reliable Reinforcement Learning
137. NeurIPS 2026CS-DICE: Offline Reinforcement Learning with Coherent Occupancy Regularization
138. NeurIPS 2026CURe: Conservative Unlearning with Soft-Gating Regularization for Offline Reinforcement Learning
139. NeurIPS 2026CURE: Coupled User-Grouped Reinforcement Learning for Cross-Domain Recommendation with Non-Overlapping Users
140. NeurIPS 2026Curriculum proof repair with learned counterexamples
141. NeurIPS 2026CurveRL: Principled Distribution-Aware Context Reweighting for LLM Reasoning
142. NeurIPS 2026CyCLeGen: Cycle-Consistent Layout Prediction and Image Generation
143. NeurIPS 2026D-VLA: A High-Concurrency Distributed Asynchronous RL Framework for Vision-Language-Action Models
144. NeurIPS 2026DACA-GRPO: Denoising-Aware Credit Assignment for Reinforcement Learning in Diffusion Language Models
145. NeurIPS 2026DACE RL for Compute Efficient Reinforcement Learning in Small Model Reasoning
146. NeurIPS 2026DaCe-DT: Data-Centric Offline Multi-Task Reinforcement Learning via Adaptive Prompts and Trajectory Correction for Heterogeneous Tasks
147. NeurIPS 2026DARLING: Detection Augmented Reinforcement Learning with Non-Stationary Guarantees
148. NeurIPS 2026DASM: Dynamic Autoregressive Subgraph Mining via Two-Stage Policy Alignment
149. NeurIPS 2026DDACOM: Dual-Decoupled Adaptive Communication for Multi-Agent Reinforcement Learning under Dynamic Networks
150. NeurIPS 2026Deadline-Constrained Dynamic Workflow Scheduling Can be Cast as a Representation Learning Problem

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。