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 911 篇，第 17 / 19 页

801. ICLR 2026TikZilla: Scaling Text-to-TikZ with High-Quality Data and Reinforcement Learning
802. ICLR 2026Time Is a Feature: Exploiting Temporal Dynamics in Diffusion Language Models
803. ICLR 2026TimeSearch-R: Adaptive Temporal Search for Long-Form Video Understanding via Self-Verification Reinforcement Learning
804. ICLR 2026TimeSeg: An Information-Theoretic Segment-Wise Explainer for Time-Series Predictions
805. ICLR 2026Tina: Tiny Reasoning Models via LoRA
806. ICLR 2026TIPO: Text to Image with Text Pre-sampling for Prompt Optimization
807. ICLR 2026TIPS: Turn-level Information-Potential Reward Shaping for Search-Augmented LLMs
808. ICLR 2026Token Hidden Reward: Steering Exploration-Exploitation in Group Relative Deep Reinforcement Learning
809. ICLR 2026Token-Efficient Long-Term Interest Sketching and Internalized Reasoning for LLM-based Recommendation
810. ICLR 2026Token-Guard: Towards Token-Level Hallucination Control via Self-Checking Decoding
811. ICLR 2026Toward Conservative Planning from Preferences in Offline Reinforcement Learning
812. ICLR 2026Toward Efficient Exploration by Large Language Model Agents
813. ICLR 2026Towards a Sharp Analysis of Learning Offline $f$-Divergence-Regularized Contextual Bandits
814. ICLR 2026Towards Better Optimization For Listwise Preference in Diffusion Models
815. ICLR 2026Towards Bridging the Gap between Large-Scale Pretraining and Efficient Finetuning for Humanoid Control
816. ICLR 2026Towards Efficient Constraint Handling in Neural Solvers for Routing Problems
817. ICLR 2026Towards High Data Efficiency in Reinforcement Learning with Verifiable Reward
818. ICLR 2026Towards Improvisational TAMP: Learning Low-Level Shortcuts in Abstract Planning Graphs
819. ICLR 2026Towards Strategic Persuasion with Language Models
820. ICLR 2026Towards Sustainable Investment Policies Informed by Opponent Shaping
821. ICLR 2026TPRU: Advancing Temporal and Procedural Understanding in Large Multimodal Models
822. ICLR 2026Traceable Evidence Enhanced Visual Grounded Reasoning: Evaluation and Method
823. ICLR 2026TRACED: Transition-aware Regret Approximation with Co-learnability for Environment Design
824. ICLR 2026Training Large Reasoning Models Efficiently via Progressive Thought Encoding
825. ICLR 2026Trajectory Generation with Conservative Value Guidance for Offline Reinforcement Learning
826. ICLR 2026Transitive RL: Value Learning via Divide and Conquer
827. ICLR 2026Translate Policy to Language: Flow Matching Generated Rewards for LLM Explanations
828. ICLR 2026TraPO: A Semi-Supervised Reinforcement Learning Framework for Boosting LLM Reasoning
829. ICLR 2026Tree Search for LLM Agent Reinforcement Learning
830. ICLR 2026Tree-based Dialogue Reinforced Policy Optimization for Red-Teaming Attacks
831. ICLR 2026TreeGRPO: Tree-Advantage GRPO for Online RL Post-Training of Diffusion Models
832. ICLR 2026Tricks or Traps? A Deep Dive into RL for LLM Reasoning
833. ICLR 2026TRIM: Hybrid Inference via Targeted Stepwise Routing in Multi-Step Reasoning Tasks
834. ICLR 2026Triple-BERT: Do We Really Need MARL for Order Dispatch on Ride-Sharing Platforms?
835. ICLR 2026TrojanTO: Action-Level Backdoor Attacks Against Trajectory Optimization Models
836. ICLR 2026TROLL: Trust Regions Improve Reinforcement Learning for Large Language Models
837. ICLR 2026Trust-Region Adaptive Policy Optimization
838. ICLR 2026TTS Can Speak in Any Style with Any Voice
839. ICLR 2026UI-Ins: Enhancing GUI Grounding with Multi-Perspective Instruction as Reasoning
840. ICLR 2026UME-R1: Exploring Reasoning-Driven Generative Multimodal Embeddings
841. ICLR 2026Uncovering Robot Vulnerabilities through Semantic Potential Fields
842. ICLR 2026Understanding and Improving Hyperbolic Deep Reinforcement Learning
843. ICLR 2026Understanding VLMs Spatial Mental Modeling Capability from Limited Views
844. ICLR 2026Uni-DPO: A Unified Paradigm for Dynamic Preference Optimization of LLMs
845. ICLR 2026Unified Biomolecular Trajectory Generation via Pretrained Variational Bridge
846. ICLR 2026Unifying Stable Optimization and Reference Regularization in RLHF
847. ICLR 2026UNIVERSAL AND EFFICIENT LOADING BALANCING FOR RL TRAINING OF LARGE MULTIMODAL MODELS
848. ICLR 2026Universal Value-Function Uncertainties
849. ICLR 2026Unleashing Perception-Time Scaling to Multimodal Reasoning Models
850. ICLR 2026Unleashing Scientific Reasoning for Bio-experimental Protocol Generation via Structured Component-based Reward Mechanism

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。