代码生成论文趋势 2023–2026

主要结论

• 代码生成论文从 2023 年的 210 篇增长到 2026 年的 860 篇，三年间扩大到约 4.1 倍。
• NeurIPS 是发代码生成论文最多的会议，2023–2026 年共 630 篇，占四个会议总量的 29%。
• 交叉方向里涨得最快的是AI 智能体：同时属于两个主题的论文从 2023 年的 0 篇增至 2026 年的 93 篇。
• 在「自然语言」的 5 个主题里，它的 2026 年论文数排第 1；同组增长最快的检索增强生成是 5.4 倍，代码生成是 4.1 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 Theoretical Analysis of Test-Driven Code Generation
• ICML 2026$A_2$DEPT: Large Language Model–Driven Automated Algorithm Design via Evolutionary Program Trees
• ICLR 2026A Bayesian Nonparametric Framework for Private, Fair, and Balanced Tabular Data Synthesis
• CVPR 2026ACPV-Net All-Class Polygonal Vectorization for Seamless Vector Map Generation from
• NeurIPS 2025$\texttt{STRCMP}$: Integrating Graph Structural Priors with Language Models for Combinatorial Optimization
• ICML 2025AlphaVerus: Bootstrapping Formally Verified Code Generation through Self-Improving Translation and Treefinement
• ICLR 2025Automated Proof Generation for Rust Code via Self-Evolution
• CVPR 2025ArtFormer: Controllable Generation of Diverse 3D Articulated Objects
• NeurIPS 2024AMBROSIA: A Benchmark for Parsing Ambiguous Questions into Database Queries
• ICML 2024ACM-MILP: Adaptive Constraint Modification via Grouping and Selection for Hardness-Preserving MILP Instance Generation
• ICLR 2024CODE REPRESENTATION LEARNING AT SCALE
• CVPR 2024Desigen: A Pipeline for Controllable Design Template Generation
• NeurIPS 2023ANPL: Towards Natural Programming with Interactive Decomposition
• ICML 2023Can Large Language Models Reason about Program Invariants?
• ICLR 2023Binding Language Models in Symbolic Languages
• CVPR 2023Attribute-Preserving Face Dataset Anonymization via Latent Code Optimization
• NeurIPS 2026A Theory of Time-Sensitive Language Generation: Sparse Hallucination Beats Mode Collapse
• ICML 2026Beyond Problem Solving: UOJ-Bench for Evaluating Code Generation, Hacking, and Repair in Competitive Programming
• ICLR 2026AssetFormer: Modular 3D Assets Generation with Autoregressive Transformer
• CVPR 2026Animator-Centric Skeleton Generation on Objects with Fine-Grained Details

全部代码生成论文

共 860 篇，第 17 / 18 页

801. NeurIPS 2026Representation Fréchet Loss for Visual Generation
802. NeurIPS 2026Rethinking Personalized Generation: Test-time Alignment via Factorized Ranking Models
803. NeurIPS 2026RevengeBench: Reverse Engineering Code-Space Policies from Behavioral Experiments
804. NeurIPS 2026Reward Shaping to Improve Language Model Query Generation
805. NeurIPS 2026RoboExo: Structure-Guided Wrist-to-Exocentric Video Generation for Scalable Robot Learning
806. NeurIPS 2026s2n-bignum-bench: A practical benchmark for evaluating low-level code reasoning of LLMs
807. NeurIPS 2026Safe Few-Step Generation via Velocity Editing
808. NeurIPS 2026SAGE: Semantic Ambiguity Guided Capacity Expansion for Retrieval-Augmented Generation
809. NeurIPS 2026SDFlow: Similarity-Driven Flow Matching for Time Series Generation
810. NeurIPS 2026Selected-Tail Reliability in Verifier-Guided Best-of-$N$ Inference
811. NeurIPS 2026SemGeo-Gen: Unsupervised Generation of Approximate Cross-Instance Semantic-Geometric Correspondences
812. NeurIPS 2026SemISP: Semantic-Consistent Diffusion for Cross-Camera RAW-to-sRGB Generation
813. NeurIPS 2026SILSA: Sliding-Window Slice Latents for Topology-Preserving High-Resolution 3D Generation
814. NeurIPS 2026Simple Baselines are Competitive with Code Evolution
815. NeurIPS 2026Simple Test-Time Refinement for Plot-to-Code Generation via Visual-Code Diagnostics
816. NeurIPS 2026SimSD: Simple Speculative Decoding in Diffusion Language Models
817. NeurIPS 2026SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learning
818. NeurIPS 2026SliceWorld: A Predictive and Controllable World-State Model for CT Report Generation
819. NeurIPS 2026Source-Causal Control of Historical Context in Longitudinal Radiology Report Generation
820. NeurIPS 2026Spatially Feasible 3D Indoor Scene Generation via Interaction-Oriented Human Proxy
821. NeurIPS 2026Spherical Boltzmann machines: a solvable theory of learning and generation in energy-based models
822. NeurIPS 2026SPRING: Solver-guided Process Rewards for Novel Logical Reasoning Steps Generation
823. NeurIPS 2026Steering Fields: Adaptive Vector Fields for Safe Image Generation and Beyond
824. NeurIPS 2026StepCAD: Mesh-to-CAD Code Generation via LLM Policy and Geometry-Guided Search
825. NeurIPS 2026Structural Self-Teaching for Compositional Generation in Unified Multimodal Models
826. NeurIPS 2026SwitchLingua V2: Agent-Driven Code-Switching via Digital Clones
827. NeurIPS 2026SynBench: A Benchmark for Differentially Private Text Generation
828. NeurIPS 2026SynMQG: Disentanglement and Mutual-Information Optimization for Synergistic Multi-modal Question Generation
829. NeurIPS 2026Synthon Contrastive Learning for Synthesizable 3D Molecule Generation
830. NeurIPS 2026TaxaAdapter: Scaling Fine-grained Species Image Generation To the Tree of Life
831. NeurIPS 2026TGPO: Trace-Guided Policy Optimization for Robot Task Planning via Verifiable Subgoal Generation
832. NeurIPS 2026The Future of Facts: Tracing the Factual Generation-Verification Gap
833. NeurIPS 2026The Sharp Directions Are Against You: Curvature Analysis of Activation Steering
834. NeurIPS 2026The Sign Code: The Hidden Binary Nature of Deep Networks
835. NeurIPS 2026The Unembedding Bottleneck: A Mechanistic Analysis of Single Digit Counting in LLMs
836. NeurIPS 2026TimeTok: Granularity-Controllable Time-Series Generation via Hierarchical Tokenization
837. NeurIPS 2026TKCAM: Text and Keyframe to Camera Trajectory Generation
838. NeurIPS 2026Tokenizer-Generator Coupling in Medical Image Generation
839. NeurIPS 2026Toward Executable Multi-framework Front-end Code Generation with Self-Correction
840. NeurIPS 2026Toward Interactive Understanding of Code APIs
841. NeurIPS 2026Towards Direct Latent-Space Synthesis for Parallel Branches in LLM-Agent Workflows
842. NeurIPS 2026TransitLM: A Large-Scale Dataset and Benchmark for Map-Free Transit Route Generation
843. NeurIPS 2026Trellis4D: Complex 4D Mesh Generation
844. NeurIPS 2026UniPath: Adaptive Coordination of Understanding and Generation for Unified Multimodal Reasoning
845. NeurIPS 2026UniSHARP: Universal Sharp Monocular View Synthesis
846. NeurIPS 2026Useful Memories Become Faulty When Continuously Updated by LLMs
847. NeurIPS 2026Veri-Sure: Multi-Agent RTL Code Generation with Temporal Tracing, Slicing and Formal Verification
848. NeurIPS 2026VeriContest: A Competitive-Programming Benchmark for Verifiable Code Generation
849. NeurIPS 2026Verification-Guided Abstraction Generation with Large Language Models for Generalized Planning
850. NeurIPS 2026Verifier-Backed Hard Problem Generation for Mathematical Reasoning

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。