代码生成论文趋势 2023–2026

程序合成与代码 LLM

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 2,161 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

代码生成论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
202329553195210
202413566111134446
202523016479172645
2026236241174209860

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

全部代码生成论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 236 篇，第 4 / 5 页

  1. NeurIPS 2026Orthrus: Memory-Efficient Parallel Token Generation via Dual-View Diffusion
  2. NeurIPS 2026OverLay++: Dense-Overlap Layout-to-Image Generation Dataset
  3. NeurIPS 2026P-EAGLE: Parallel-Drafting EAGLE with Scalable Training
  4. NeurIPS 2026P$^{3}$: Joint Program-and-Proof Planning\\ for Verified Code Generation
  5. NeurIPS 2026Panoptic Scene Program Diffusion Transformer
  6. NeurIPS 2026Parallel Rollout Approximation for Pixel-Space Autoregressive Image Generation
  7. NeurIPS 2026PBT-Bench: Benchmarking AI Agents on Property-Based Testing
  8. NeurIPS 2026Permit: Permission-Aware Representation Intervention for Controlled Generation in Large Language Models
  9. NeurIPS 2026Phase-wise MLLM Tuning for Multi-framework WebUI Code Generation
  10. NeurIPS 2026PhysAgentGym: Free Physical-Law Verifiers for Training Small Code-Reasoning Agents
  11. NeurIPS 2026Population-Aligned Persona Generation for LLM-based Social Simulation
  12. NeurIPS 2026Prefix Likelihood-Ratio Control: Tail-Stable Training for Long-Horizon Language Generation
  13. NeurIPS 2026Primal Generation, Dual Judgment: Self-Training from Test-Time Scaling
  14. NeurIPS 2026Primal-Dual Flow Matching for Sample-Wise Constrained Generation
  15. NeurIPS 2026PRO: Enabling Precise and Robust Text Watermark for Open-Source LLMs
  16. NeurIPS 2026Property-Guided LLM Program Synthesis for Planning
  17. NeurIPS 2026PSD: Pushing the Pareto Frontier of Diffusion LLMs via Parallel Speculative Decoding
  18. NeurIPS 2026RAVEL: Rare Concept Generation and Editing via Graph-driven Relational Guidance
  19. NeurIPS 2026Re-evaluating Confidence Remasking in Masked Diffusion Language Models
  20. NeurIPS 2026Recursive Language Models
  21. NeurIPS 2026Refactoring Code Through Library Design
  22. NeurIPS 2026REFLEX: RNA Ensemble Generation via Flexibility-Calibrated Stochastic Bridge
  23. NeurIPS 2026Reinforcement Learning for Code Optimization
  24. NeurIPS 2026Reinforcement Learning with Verifiable Physics: Post-training LLMs for PDE Solver Generation
  25. NeurIPS 2026Render Structure Uncertainty for HTML Repair in MLLM-based UI-to-Code Generation
  26. NeurIPS 2026RepoZero: Can LLMs Generate a Code Repository from Scratch?
  27. NeurIPS 2026Representation Fréchet Loss for Visual Generation
  28. NeurIPS 2026Rethinking Personalized Generation: Test-time Alignment via Factorized Ranking Models
  29. NeurIPS 2026RevengeBench: Reverse Engineering Code-Space Policies from Behavioral Experiments
  30. NeurIPS 2026Reward Shaping to Improve Language Model Query Generation
  31. NeurIPS 2026RoboExo: Structure-Guided Wrist-to-Exocentric Video Generation for Scalable Robot Learning
  32. NeurIPS 2026s2n-bignum-bench: A practical benchmark for evaluating low-level code reasoning of LLMs
  33. NeurIPS 2026Safe Few-Step Generation via Velocity Editing
  34. NeurIPS 2026SAGE: Semantic Ambiguity Guided Capacity Expansion for Retrieval-Augmented Generation
  35. NeurIPS 2026SDFlow: Similarity-Driven Flow Matching for Time Series Generation
  36. NeurIPS 2026Selected-Tail Reliability in Verifier-Guided Best-of-$N$ Inference
  37. NeurIPS 2026SemGeo-Gen: Unsupervised Generation of Approximate Cross-Instance Semantic-Geometric Correspondences
  38. NeurIPS 2026SemISP: Semantic-Consistent Diffusion for Cross-Camera RAW-to-sRGB Generation
  39. NeurIPS 2026SILSA: Sliding-Window Slice Latents for Topology-Preserving High-Resolution 3D Generation
  40. NeurIPS 2026Simple Baselines are Competitive with Code Evolution
  41. NeurIPS 2026Simple Test-Time Refinement for Plot-to-Code Generation via Visual-Code Diagnostics
  42. NeurIPS 2026SimSD: Simple Speculative Decoding in Diffusion Language Models
  43. NeurIPS 2026SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learning
  44. NeurIPS 2026SliceWorld: A Predictive and Controllable World-State Model for CT Report Generation
  45. NeurIPS 2026Source-Causal Control of Historical Context in Longitudinal Radiology Report Generation
  46. NeurIPS 2026Spatially Feasible 3D Indoor Scene Generation via Interaction-Oriented Human Proxy
  47. NeurIPS 2026Spherical Boltzmann machines: a solvable theory of learning and generation in energy-based models
  48. NeurIPS 2026SPRING: Solver-guided Process Rewards for Novel Logical Reasoning Steps Generation
  49. NeurIPS 2026Steering Fields: Adaptive Vector Fields for Safe Image Generation and Beyond
  50. NeurIPS 2026StepCAD: Mesh-to-CAD Code Generation via LLM Policy and Geometry-Guided Search

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 查代码生成在 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": "code-generation"}

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代码生成论文？
共 2,161 篇，按年份分别是 2023 年 210 篇、2024 年 446 篇、2025 年 645 篇、2026 年 860 篇。
NeurIPS 2026 有多少篇代码生成论文？
NeurIPS 2026 有 236 篇代码生成论文，上一届是 230 篇。
ICLR 2026 有多少篇代码生成论文？
ICLR 2026 有 241 篇代码生成论文，上一届是 164 篇。
ICML 2026 有多少篇代码生成论文？
ICML 2026 有 174 篇代码生成论文，上一届是 79 篇。
CVPR 2026 有多少篇代码生成论文？
CVPR 2026 有 209 篇代码生成论文，上一届是 172 篇。
哪个会议的代码生成论文最多？
NeurIPS，2023–2026 年共 630 篇。

「自然语言」下的其他主题