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 1,983 篇，第 34 / 40 页

  1. NeurIPS 2026Mental-Models for Multi-Agent Systems
  2. NeurIPS 2026Meta-Cognitive Memory Policy Optimization for Long-Horizon LLM Agents
  3. NeurIPS 2026Meta$^n$: Recursive Self-Improvement through Emergent Depth
  4. NeurIPS 2026MGMem: An Efficient, Deterministic, and Provenance-Preserving Framework for Long-Horizon Agent Memory
  5. NeurIPS 2026Mind the Gap: Information Disadvantage as a Learning Signal in Cooperative MARL
  6. NeurIPS 2026MindGames:A Multi-Agent Benchmark and Trajectory Dataset for Evaluating Social and Strategic Reasoning in LLMs
  7. NeurIPS 2026MineEvolve: Self-Evolution with Accumulated Knowledge for Long-Horizon Embodied Minecraft Agents
  8. NeurIPS 2026MiniPIC: Flexible Position-Independent Caching in <100LOC
  9. NeurIPS 2026Mirror, Mirror on the Wall: Can VLM Agents Tell Who They Are at All?
  10. NeurIPS 2026MLLMs Fail to Refuse when Using Tools Agentically
  11. NeurIPS 2026MLS-Bench: A Holistic and Rigorous Assessment of AI Systems on Building Better AI
  12. NeurIPS 2026MMGraph-Agent: Agentic Multimodal RAG via Cache-Inspired Multimodal Knowledge HyperGraphs
  13. NeurIPS 2026MMSkills: Towards Multimodal Skills for General Visual Agents
  14. NeurIPS 2026Mode-Controlled Policy Optimization: A Geometry-Aware Recipe for LLM Post-Training
  15. NeurIPS 2026Model Incrimination: Investigating Whether Concerning Behavior Reflects Misalignment
  16. NeurIPS 2026Model-Adaptive Tool Necessity Reveals the Knowing-Doing Gap in LLM Tool Use
  17. NeurIPS 2026Modelling Opinion Dynamics at Scale with Deep MARL
  18. NeurIPS 2026Models Recall What They Violate: Constraint Adherence in Multi-Turn LLM Ideation
  19. NeurIPS 2026MoMHa: Multi-Objective Optimization of LLM Harnesses over Accuracy, Safety, and Tokens
  20. NeurIPS 2026More Than Meets the Eye? Uncovering the Reasoning-Planning Disconnect in Training Vision-Language Driving Models
  21. NeurIPS 2026More Value per Key: Asymmetric Sparse Attention for Faster LLM Decoding
  22. NeurIPS 2026MOSAIC-Bench: Measuring Compositional Vulnerability Induction in Coding Agents
  23. NeurIPS 2026MOSAIC: Scaling Long-Horizon Language Agents via Multi-Scale Adaptive Inference Control
  24. NeurIPS 2026Motion Forcing: Decoupling Ego and Object Motion via Sparse Inputs for Structured Video Generation
  25. NeurIPS 2026MPCI-Bench: A Benchmark for Multimodal Pairwise Contextual Integrity Privacy Evaluation of Language Model Agents
  26. NeurIPS 2026Multi-agent Collaboration with State Management
  27. NeurIPS 2026Multi-Agent Coordination via Support-Preserving Distillation
  28. NeurIPS 2026Multi-bit LLM Watermarking with Certified Semantic Distortion
  29. NeurIPS 2026MultiSTEVE-1s: A Model Zoo and Interpretability Suite for Instruction-Following Vision Agents
  30. NeurIPS 2026NARRA-Gym for Evaluating Interactive Narrative Agents
  31. NeurIPS 2026Narrowing the Collaboration Gap, Probably
  32. NeurIPS 2026Nautilus: From One Prompt to Plug-and-Play Robot Learning
  33. NeurIPS 2026Need-Aware Multi-Objective Reinforcement Learning for Emotionally Intelligent LLM Agents
  34. NeurIPS 2026Neglected Free Lunch from Post-training: Progress Advantage for LLM Agents
  35. NeurIPS 2026NEST: Nascent Encoded Steganographic Thoughts
  36. NeurIPS 2026NestRL: A Nested Training Regime for Mutual Adaptation in Human–AI Teaming
  37. NeurIPS 2026NeSyKC: Neurosymbolic Knowledge Compilation For Lifelong Learning Embodied Agents
  38. NeurIPS 2026NesyProAct: Proactive Neural-Symbolic Control for Web Agents
  39. NeurIPS 2026NINJA: A Navigator–Inspector Joint Architecture for Context-Efficient Issue Localization
  40. NeurIPS 2026NitroBox: Lightning-Fast Sandbox for Large-Scale RL Training
  41. NeurIPS 2026No Model Required: Text Entropy Rate Filtering Prevents Iterative Fine-Tuning Collapse
  42. NeurIPS 2026Norm Enforcement for AI Agents: Robustly Shaping Behavior in Multi-Agent Systems
  43. NeurIPS 2026Not All Proofs Are Equal: Evaluating LLM Proof Quality Beyond Correctness
  44. NeurIPS 2026ODDR: One-Step Deshadow Diffusion via Reward Guidance
  45. NeurIPS 2026OfficeInstruct: A Dataset for Tool-Augmented Office Agents on Long-Trajectory Generative Tasks
  46. NeurIPS 2026Offline Preference-Based Trajectory Evaluation
  47. NeurIPS 2026On Data Engineering for Scaling LLM Terminal Capabilities
  48. NeurIPS 2026On the Impact of Side-Information in Bandit Learning: More is Not Always Merrier
  49. NeurIPS 2026On the Nonlinearity of Learning Rate Scaling for LLM Training
  50. NeurIPS 2026One Adapter For All: Towards Generalizable Many-To-One Domain Adaptation In Heterogeneous Collaborative Perception

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