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 645 篇，第 9 / 13 页

  1. ICML 2025Revisiting Chain-of-Thought in Code Generation: Do Language Models Need to Learn Reasoning before Coding?
  2. ICML 2025Scalable Non-Equivariant 3D Molecule Generation via Rotational Alignment
  3. ICML 2025Selective Prompt Anchoring for Code Generation
  4. ICML 2025Self-Improving Language Models for Evolutionary Program Synthesis: A Case Study on ARC-AGI
  5. ICML 2025SongGen: A Single Stage Auto-regressive Transformer for Text-to-Song Generation
  6. ICML 2025SpeCache: Speculative Key-Value Caching for Efficient Generation of LLMs
  7. ICML 2025Strategic Planning: A Top-Down Approach to Option Generation
  8. ICML 2025Synthesizing Software Engineering Data in a Test-Driven Manner
  9. ICML 2025The Diffusion Duality
  10. ICML 2025Thinking LLMs: General Instruction Following with Thought Generation
  11. ICML 2025TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation
  12. ICML 2025Towards Practical Defect-Focused Automated Code Review
  13. ICML 2025Tractable Transformers for Flexible Conditional Generation
  14. ICML 2025TransPL: VQ-Code Transition Matrices for Pseudo-Labeling of Time Series Unsupervised Domain Adaptation
  15. ICML 2025Understanding Complexity in VideoQA via Visual Program Generation
  16. NeurIPS 2025$\texttt{STRCMP}$: Integrating Graph Structural Priors with Language Models for Combinatorial Optimization
  17. NeurIPS 2025AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforced Self-Play
  18. NeurIPS 2025Activation-Informed Merging of Large Language Models
  19. NeurIPS 2025Adaptive 3D Reconstruction via Diffusion Priors and Forward Curvature-Matching Likelihood Updates
  20. NeurIPS 2025Adaptive Discretization for Consistency Models
  21. NeurIPS 2025Advanced Sign Language Video Generation with Compressed and Quantized Multi-Condition Tokenization
  22. NeurIPS 2025AdvPrefix: An Objective for Nuanced LLM Jailbreaks
  23. NeurIPS 2025Afterburner: Reinforcement Learning Facilitates Self-Improving Code Efficiency Optimization
  24. NeurIPS 2025AI Research Agents for Machine Learning: Search, Exploration, and Generalization in MLE-bench
  25. NeurIPS 2025AlignedGen: Aligning Style Across Generated Images
  26. NeurIPS 2025AlphaDecay: Module-wise Weight Decay for Heavy-Tailed Balancing in LLMs
  27. NeurIPS 2025ALTo: Adaptive-Length Tokenizer for Autoregressive Mask Generation
  28. NeurIPS 2025AnimateQR: Bridging Aesthetics and Functionality in Dynamic QR Code Generation
  29. NeurIPS 2025APML: Adaptive Probabilistic Matching Loss for Robust 3D Point Cloud Reconstruction
  30. NeurIPS 2025Approximately Aligned Decoding
  31. NeurIPS 2025AstroVisBench: A Code Benchmark for Scientific Computing and Visualization in Astronomy
  32. NeurIPS 2025ATLAS: Autoformalizing Theorems through Lifting, Augmentation, and Synthesis of Data
  33. NeurIPS 2025AutoPartGen: Autoregressive 3D Part Generation and Discovery
  34. NeurIPS 2025BADiff: Bandwidth Adaptive Diffusion Model
  35. NeurIPS 2025Benchmarking Large Language Models with Integer Sequence Generation Tasks
  36. NeurIPS 2025Benford’s Curse: Tracing Digit Bias to Numerical Hallucination in LLMs
  37. NeurIPS 2025Bi-Directional Communication-Efficient Stochastic FL via Remote Source Generation
  38. NeurIPS 2025Bidirectional Motion Transformer for Safety-Critical Traffic Scenario Generation
  39. NeurIPS 2025Bidirectional Representations Augmented Autoregressive Biological Sequence Generation: Application in De Novo Peptide Sequencing
  40. NeurIPS 2025BIPNN: Learning to Solve Binary Integer Programming via Hypergraph Neural Networks
  41. NeurIPS 2025Blockwise Flow Matching: Improving Flow Matching Models For Efficient High-Quality Generation
  42. NeurIPS 2025BoltzNCE: Learning likelihoods for Boltzmann Generation with Stochastic Interpolants and Noise Contrastive Estimation
  43. NeurIPS 2025Boosting Knowledge Utilization in Multimodal Large Language Models via Adaptive Logits Fusion and Attention Reallocation
  44. NeurIPS 2025Brain network science modelling of sparse neural networks enables Transformers and LLMs to perform as fully connected
  45. NeurIPS 2025Bridging Sign and Spoken Languages: Pseudo Gloss Generation for Sign Language Translation
  46. NeurIPS 2025Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs
  47. NeurIPS 2025Classical Planning with LLM-Generated Heuristics: Challenging the State of the Art with Python Code
  48. NeurIPS 2025CLEAR: Conv-Like Linearization Revs Pre-Trained Diffusion Transformers Up
  49. NeurIPS 2025CLEVER: A Curated Benchmark for Formally Verified Code Generation
  50. NeurIPS 2025CodeAssistBench (CAB): Dataset & Benchmarking for Multi-turn Chat-Based Code Assistance

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