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 630 篇，第 3 / 13 页

101. NeurIPS 2026How LLMs Distinguish Threats from Offers
102. NeurIPS 2026How Much is Left? LLMs Linearly Encode Their Remaining Output Length
103. NeurIPS 2026Hypergraph Generation with Latent Diffusion
104. NeurIPS 2026HyperNSDE: Personalized Neural SDEs for Joint Static—Longitudinal Clinical Data Generation
105. NeurIPS 2026Hypothesis generation and updating in large language models
106. NeurIPS 2026IFDECORATOR: Wrapping Instruction Following Reinforcement Learning with Verifiable Rewards
107. NeurIPS 2026Imagine Before You Draw: Visual Prompt Engineering for Image Generation
108. NeurIPS 2026In-Context Optimization for Retrieval-Augmented Generation: A Gradient-Descent Perspective
109. NeurIPS 2026IndustryCode: A Benchmark for Industry Code Generation
110. NeurIPS 2026Inference-Time Vulnerability Beyond Shallow Safety: Alignment Along Generation Trajectories
111. NeurIPS 2026Information Parity for Code: The Scope of Transfer in Multilingual Code Models
112. NeurIPS 2026Instantiation of Human Values in Image Generation
113. NeurIPS 2026IntegrityBench: Can LLMs Be Trusted as Co-Scientists? A Research Integrity Benchmark
114. NeurIPS 2026Intent Factored Generation: Unleashing the Diversity in Your Language Model
115. NeurIPS 2026Intent2CAD: How Semantic-Parametric Supervision Shapes Text-to-CAD Generation
116. NeurIPS 2026Invisible Ink, Visible Lies: How Production Watermarking Causes LLMs to Hallucinate
117. NeurIPS 2026JEDI: Real-Time Jailbreak Defense for LLMs via In-Generation Detection and Intervention
118. NeurIPS 2026LADDERS: Length-Aware Data Distribution and Existing-Response Speculation for Fast RL Rollout Generation
119. NeurIPS 2026Layout Before Pixels: Topology-Anchored Transcriptome-to-Histology Generation
120. NeurIPS 2026LayoutBridge: Anisotropic Brownian Bridges for Public Indoor Floorplan Generation
121. NeurIPS 2026LEAP: Library-driven Evolutionary Abstraction Paradigm for Large Language Models
122. NeurIPS 2026Learning Cultural Vectors for Cross-Cultural Generation
123. NeurIPS 2026Learning to Persuade Exposes How Easily LLMs Abandon Correct Beliefs
124. NeurIPS 2026LibEvoBench: Probing Temporal Knowledge Stratification in Code Generation Models
125. NeurIPS 2026LiFT: Lifted Inter-slice Feature Trajectories for 3D Image Generation from 2D Generators
126. NeurIPS 2026LLMs Keep Thinking When Told Not To
127. NeurIPS 2026LLMs Optimizing LLMs: Automated MegaKernel Generation for Inference Acceleration
128. NeurIPS 2026LLMs Show No Signs Of Individuated Metacognition
129. NeurIPS 2026LoopNav: Benchmarking Spatial Consistency in World Models
130. NeurIPS 2026LoopWeaver: Weaving Feedback Loops into Hierarchical Generation under Constraints
131. NeurIPS 2026Machine Unlearning in Diffusion LLMs
132. NeurIPS 2026Matchability-Aware Conformal Prediction for Open-Ended Language Model Generation
133. NeurIPS 2026MDMR-Bench: A Multi-Dimensional Benchmark for Multi-Reference Image Generation
134. NeurIPS 2026Measuring Layer-wise Intrinsic Dimensionality of FFNs in LLMs via PCA
135. NeurIPS 2026MedIGen: Reliable Medical Illustration Generation via Interleaved Introspective Reasoning
136. NeurIPS 2026Mesh BDF: Barycentric Dominance Field for 3D Native Mesh Generation
137. NeurIPS 2026MetaLoop: Benchmarking the Full Metacognitive Loop in LLMs
138. NeurIPS 2026MobTA: Bus-Conditioned Zero-Shot Trajectory Generation via Task Arithmetic
139. NeurIPS 2026MoCA: Mixture-of-Components Attention for Scalable Compositional 3D Generation
140. NeurIPS 2026MoCAR: Motion-code Coordinate-aware AutoRegression for Continuous Trajectory Forecasting
141. NeurIPS 2026Motion Forcing: Decoupling Ego and Object Motion via Sparse Inputs for Structured Video Generation
142. NeurIPS 2026N-vium: Mixture-of-Exits Transformer for Accelerated Exact Generation
143. NeurIPS 2026Natural-Language-Guided Protein Generation for Ligand-Binding Design
144. NeurIPS 2026NeuroSynTheos: Learning to accelerate counterexample-guided reactive synthesis modulo theories
145. NeurIPS 2026NIV: Neural Axis Variations for Variable Font Generation
146. NeurIPS 2026Norm Anchors Make Model Edits Last
147. NeurIPS 2026On the Price of Privacy for Language Identification and Generation
148. NeurIPS 2026On-Policy Distillation with Open Property-Equivalence Reward for LLM-Based NL-to-SVA Generation
149. NeurIPS 2026One View Is Enough: In-the-Wild Monocular Pretraining for Novel View Generation
150. NeurIPS 2026OopsWorld! Operation-Grounded Seamless World Generation

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。