AI 智能体论文趋势 2023–2026

工具调用、多步决策与自主 LLM/VLM 智能体系统

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 3,266 篇被归入该主题的论文。

主要结论

每年论文数

各会议论文数

按年份和会议统计

AI 智能体论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
202376404113170
2024150738818329
202536220417840784
20267295006401141,983

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

全部AI 智能体论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 3,266 篇，第 48 / 66 页

  1. ICML 2025Position: Truly Self-Improving Agents Require Intrinsic Metacognitive Learning
  2. ICML 2025Position: Uncertainty Quantification Needs Reassessment for Large Language Model Agents
  3. ICML 2025Position: We Need An Algorithmic Understanding of Generative AI
  4. ICML 2025Prediction-Aware Learning in Multi-Agent Systems
  5. ICML 2025Principal-Agent Bandit Games with Self-Interested and Exploratory Learning Agents
  6. ICML 2025Proactive Agents for Multi-Turn Text-to-Image Generation Under Uncertainty
  7. ICML 2025Proposer-Agent-Evaluator (PAE): Autonomous Skill Discovery For Foundation Model Internet Agents
  8. ICML 2025Provably Efficient Algorithm for Best Scoring Rule Identification in Online Principal-Agent Information Acquisition
  9. ICML 2025Prune 'n Predict: Optimizing LLM Decision-making with Conformal Prediction
  10. ICML 2025QLASS: Boosting Language Agent Inference via Q-Guided Stepwise Search
  11. ICML 2025R3DM: Enabling Role Discovery and Diversity Through Dynamics Models in Multi-agent Reinforcement Learning
  12. ICML 2025RE-Bench: Evaluating Frontier AI R&D Capabilities of Language Model Agents against Human Experts
  13. ICML 2025Reflection-Bench: Evaluating Epistemic Agency in Large Language Models
  14. ICML 2025Reidentify: Context-Aware Identity Generation for Contextual Multi-Agent Reinforcement Learning
  15. ICML 2025Reinforce LLM Reasoning through Multi-Agent Reflection
  16. ICML 2025RepoAudit: An Autonomous LLM-Agent for Repository-Level Code Auditing
  17. ICML 2025ResearchTown: Simulator of Human Research Community
  18. ICML 2025Revisiting Cooperative Off-Policy Multi-Agent Reinforcement Learning
  19. ICML 2025Risk-Sensitive Theory of Mind: Coordinating with Agents of Unknown Bias using Cumulative Prospect Theory
  20. ICML 2025RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning
  21. ICML 2025Robust Multi-Agent Reinforcement Learning with Stochastic Adversary
  22. ICML 2025Robust Multi-bit Text Watermark with LLM-based Paraphrasers
  23. ICML 2025Robust Noise Attenuation via Adaptive Pooling of Transformer Outputs
  24. ICML 2025Sable: a Performant, Efficient and Scalable Sequence Model for MARL
  25. ICML 2025SafeArena: Evaluating the Safety of Autonomous Web Agents
  26. ICML 2025Self-Play $Q$-Learners Can Provably Collude in the Iterated Prisoner's Dilemma
  27. ICML 2025ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoning
  28. ICML 2025Simultaneous Multi-Robot Motion Planning with Projected Diffusion Models
  29. ICML 2025Solving Zero-Sum Convex Markov Games
  30. ICML 2025Star Attention: Efficient LLM Inference over Long Sequences
  31. ICML 2025StealthInk: A Multi-bit and Stealthy Watermark for Large Language Models
  32. ICML 2025SWAN: SGD with Normalization and Whitening Enables Stateless LLM Training
  33. ICML 2025SWE-Lancer: Can Frontier LLMs Earn $1 Million from Real-World Freelance Software Engineering?
  34. ICML 2025SyncMind: Measuring Agent Out-of-Sync Recovery in Collaborative Software Engineering
  35. ICML 2025Synthesizing Software Engineering Data in a Test-Driven Manner
  36. ICML 2025The Berkeley Function Calling Leaderboard (BFCL): From Tool Use to Agentic Evaluation of Large Language Models
  37. ICML 2025The Hidden Dimensions of LLM Alignment: A Multi-Dimensional Analysis of Orthogonal Safety Directions
  38. ICML 2025The Sample Complexity of Online Strategic Decision Making with Information Asymmetry and Knowledge Transportability
  39. ICML 2025The Synergy of LLMs & RL Unlocks Offline Learning of Generalizable Language-Conditioned Policies with Low-fidelity Data
  40. ICML 2025Tokenized Bandit for LLM Decoding and Alignment
  41. ICML 2025Tool Unlearning for Tool-Augmented LLMs
  42. ICML 2025Towards Efficient Online Tuning of VLM Agents via Counterfactual Soft Reinforcement Learning
  43. ICML 2025Towards Practical Defect-Focused Automated Code Review
  44. ICML 2025Training a Generally Curious Agent
  45. ICML 2025Training Software Engineering Agents and Verifiers with SWE-Gym
  46. ICML 2025UDora: A Unified Red Teaming Framework against LLM Agents by Dynamically Hijacking Their Own Reasoning
  47. ICML 2025UI-Vision: A Desktop-centric GUI Benchmark for Visual Perception and Interaction
  48. ICML 2025Unbiased Evaluation of Large Language Models from a Causal Perspective
  49. ICML 2025Understanding Bias Reinforcement in LLM Agents Debate
  50. ICML 2025UP-VLA: A Unified Understanding and Prediction Model for Embodied Agent

在 Claude、Cursor 里直接问这个趋势

接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。

可以直接复制这句话问 AI

用 paperlist 查AI 智能体在 NeurIPS 2023–2026 的论文趋势，哪一年涨得最快？

AI 会调用的工具：get_topic_stats {"topic": "ai-agents"}

两分钟接入 Paperlist MCP（Claude Code、Cursor、Codex、ChatGPT）

常见问题

2023–2026 年四大会议一共有多少篇AI 智能体论文？
共 3,266 篇，按年份分别是 2023 年 170 篇、2024 年 329 篇、2025 年 784 篇、2026 年 1,983 篇。
NeurIPS 2026 有多少篇AI 智能体论文？
NeurIPS 2026 有 729 篇AI 智能体论文，上一届是 362 篇。
ICLR 2026 有多少篇AI 智能体论文？
ICLR 2026 有 500 篇AI 智能体论文，上一届是 204 篇。
ICML 2026 有多少篇AI 智能体论文？
ICML 2026 有 640 篇AI 智能体论文，上一届是 178 篇。
CVPR 2026 有多少篇AI 智能体论文？
CVPR 2026 有 114 篇AI 智能体论文，上一届是 40 篇。
哪个会议的AI 智能体论文最多？
NeurIPS，2023–2026 年共 1,317 篇。

「前沿系统」下的其他主题