经典规划论文趋势 2023–2026

主要结论

• 经典规划论文从 2023 年的 24 篇增长到 2026 年的 79 篇，三年间扩大到约 3.3 倍。
• NeurIPS 是发经典规划论文最多的会议，2023–2026 年共 75 篇，占四个会议总量的 42%。
• 交叉方向里涨得最快的是多模态大模型：同时属于两个主题的论文从 2023 年的 0 篇增至 2026 年的 9 篇。
• 在「博弈与规划」的 5 个主题里，它的 2026 年论文数排第 4；同组增长最快的机制设计是 3.0 倍，经典规划是 3.3 倍。

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 2026A Structured LLM Framework for Inorganic Material Synthesis Planning
• ICML 2026An Exponential Separation Between Quantum and Quantum-Inspired Classical Algorithms for Linear Systems
• ICLR 2026GTool: Graph Enhanced Tool Planning with Large Language Model
• CVPR 2026HQC-NBV A Hybrid Quantum-Classical View Planning Approach
• NeurIPS 2025\(\varepsilon\)-Optimally Solving Two-Player Zero-Sum POSGs
• ICML 2025Goal-Space Planning with Subgoal Models
• ICLR 2025From an LLM Swarm to a PDDL-empowered Hive: Planning Self-executed Instructions in a Multi-modal Jungle
• CVPR 2025Bridging Past and Future: End-to-End Autonomous Driving with Historical Prediction and Planning
• NeurIPS 2024WenMind: A Comprehensive Benchmark for Evaluating Large Language Models in Chinese Classical Literature and Language Arts
• ICML 2024A Bayesian Approach to Online Planning
• ICLR 2024Learning Planning Abstractions from Language
• CVPR 2024Why Not Use Your Textbook? Knowledge-Enhanced Procedure Planning of Instructional Videos
• NeurIPS 2023Experiment Planning with Function Approximation
• ICML 2023A Hybrid Quantum-Classical Approach based on the Hadamard Transform for the Convolutional Layer
• ICLR 2023Fast and Precise: Adjusting Planning Horizon with Adaptive Subgoal Search
• CVPR 2023Planning-Oriented Autonomous Driving
• NeurIPS 2026Behaving Better, Thinking Worse: Sycophancy Across Post-Training Stages
• ICML 2026On the Ability of Transformers to Verify Plans
• ICLR 2026Experience-based Knowledge Correction for Robust Planning in Minecraft
• CVPR 2026No Labels No Look-Ahead Unsupervised Online Video Stabilization with Classical

全部经典规划论文

共 40 篇，第 1 / 1 页

1. NeurIPS 2026A Structured LLM Framework for Inorganic Material Synthesis Planning
2. NeurIPS 2026Behaving Better, Thinking Worse: Sycophancy Across Post-Training Stages
3. NeurIPS 2026Beyond Expected Values: Risk-Sensitive Planning with Distributional Monte-Carlo Tree Search
4. NeurIPS 2026Coarse-to-Fine Compositional Diffusion for Long-Horizon Planning
5. NeurIPS 2026Complexity of Classical Acceleration for $\ell_1$-Regularized PageRank
6. NeurIPS 2026Corrupted Plans, Clean Traces: What Planning-Execution Decoupling Reveals About CoT Monitoring
7. NeurIPS 2026DAGent: Evaluate-then-Grow Planning for Deep Research Agents
8. NeurIPS 2026Differentiable Learning of Lifted Action Schemas for Classical Planning
9. NeurIPS 2026Diffusion Subgoal Planning for Long-Horizon Offline Goal-Conditioned Reinforcement Learning
10. NeurIPS 2026Discrete Diffusion Models Exploit Asymmetry to Solve Lookahead Planning Tasks
11. NeurIPS 2026EpicWorldModel: Exploration-driven Planning with Latent World Models
12. NeurIPS 2026Flexible Intensities Matter: A comprehensive re-evaluation of Classical and Neural Temporal Point Processes
13. NeurIPS 2026HaM-World: Soft-Hamiltonian World Models with Selective Memory for Planning
14. NeurIPS 2026Implicit-Euler Value Iteration: Long-Horizon Planning via Stable Integration of the Bellman Residual Flow
15. NeurIPS 2026Locating and Repairing Domain Shift in VLM Trajectory Planning
16. NeurIPS 2026Majority-of-Three is an Optimal PAC Learner
17. NeurIPS 2026Marginal-Nonuniform Multiclass Learning
18. NeurIPS 2026MemPlan: Memory-Conditioned PDDL Planning for Partially Observable Text Environments
19. NeurIPS 2026Multi-Stage Planning from Single-Stage Data: Reinforcement Learning Helps Composition but Requires Anchoring
20. NeurIPS 2026Nearly Optimal Fixed-Confidence Best-Arm Identification with 1-Bit Feedback
21. NeurIPS 2026ObjView-Bench: Rethinking Difficulty and Deployment for Object-Centric View Planning
22. NeurIPS 2026OntoPlan: An Ontology-Grounded Scene Representation and Agentic Framework for Scalable Robot Task Planning
23. NeurIPS 2026Optimizer-Induced Mode Connectivity: From AdamW to Muon
24. NeurIPS 2026Ordinal Geometry Complements Reconstruction: Diagnosing Planning with Compressed Value Functions
25. NeurIPS 2026Perfect Parallelization in Mini-Batch SGD with Classical Momentum Acceleration
26. NeurIPS 2026Plan2Sense: Open-World Task Planning in Epistemic States via Interleaved Ontic and Sensing Actions
27. NeurIPS 2026PO-PDDL: Learning Symbolic POMDPs from Visual Demonstrations for Robot Planning Under Uncertainty
28. NeurIPS 2026Posterior-Tracking for Best-Arm Identification in Bernoulli Bandits
29. NeurIPS 2026Probing Visual Planning in Image Editing Models
30. NeurIPS 2026Property-Guided LLM Program Synthesis for Planning
31. NeurIPS 2026Recovering the Apresjan Hierarchy Using Linkage-Based Clustering
32. NeurIPS 2026Refining Compositional Diffusion for Reliable Long-Horizon Planning
33. NeurIPS 2026Risk-Averse Online POMDP Planning via CVaR of the Immediate Cost with Performance Guarantees
34. NeurIPS 2026Settling the Sample Complexity of Deterministic Agnostic PAC Learning
35. NeurIPS 2026TGPO: Trace-Guided Policy Optimization for Robot Task Planning via Verifiable Subgoal Generation
36. NeurIPS 2026Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action
37. NeurIPS 2026Tokenizer Choice Shapes Generalization in State-Centric Learning for Planning
38. NeurIPS 2026Trajectory Planning without Trajectory Data: A Manifold-Guided Approach
39. NeurIPS 2026Unifying Reasoning and Planning through Energy Minimization
40. NeurIPS 2026Winning Lottery Tickets in Neural Networks via a Quantum-Inspired Classical Algorithm

常见问题

2023–2026 年四大会议一共有多少篇经典规划论文？

共 179 篇，按年份分别是 2023 年 24 篇、2024 年 35 篇、2025 年 41 篇、2026 年 79 篇。

NeurIPS 2026 有多少篇经典规划论文？

NeurIPS 2026 有 40 篇经典规划论文，上一届是 14 篇。

ICLR 2026 有多少篇经典规划论文？

ICLR 2026 有 17 篇经典规划论文，上一届是 10 篇。

ICML 2026 有多少篇经典规划论文？

ICML 2026 有 13 篇经典规划论文，上一届是 15 篇。

CVPR 2026 有多少篇经典规划论文？

CVPR 2026 有 9 篇经典规划论文，上一届是 2 篇。

哪个会议的经典规划论文最多？

NeurIPS，2023–2026 年共 75 篇。