符号回归论文趋势 2023–2026

主要结论

• 符号回归论文从 2023 年的 63 篇增长到 2026 年的 97 篇，三年间扩大到约 1.5 倍。
• NeurIPS 是发符号回归论文最多的会议，2023–2026 年共 118 篇，占四个会议总量的 37%。
• 交叉方向里涨得最快的是理论 · learning theory：同时属于两个主题的论文从 2023 年的 1 篇增至 2026 年的 10 篇。
• 在「程序综合」的 5 个主题里，它的 2026 年论文数排第 1；同组增长最快的定理证明是 13 倍，符号回归是 1.5 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 2026Conformal Prediction for Distribution-to-Distribution Regression
• ICML 2026Approximation Bounds for Transformer Networks with Application to Regression
• ICLR 2026Fast Estimation of Wasserstein Distances via Regression on Sliced Wasserstein Distances
• CVPR 2026CoLoR The Devil is in Scene Coordinate Regression for Large-Scale
• NeurIPS 2025Ascent Fails to Forget
• ICML 2025Benefits of Early Stopping in Gradient Descent for Overparameterized Logistic Regression
• ICLR 2025A Graph Enhanced Symbolic Discovery Framework For Efficient Logic Optimization
• CVPR 2025R-SCoRe: Revisiting Scene Coordinate Regression for Robust Large-Scale Visual Localization
• NeurIPS 2024Distributed Sparse Regression via Penalization
• ICML 2024$H$-Consistency Guarantees for Regression
• ICLR 2024Geometrically Aligned Transfer Encoder for Inductive Transfer in Regression Tasks
• CVPR 2024Map-Relative Pose Regression for Visual Re-Localization
• NeurIPS 2023Collaboratively Learning Linear Models with Structured Missing Data
• ICML 2023A Nearly-Optimal Bound for Fast Regression with $\ell_\infty$ Guarantee
• ICLR 2023Fast Nonlinear Vector Quantile Regression
• CVPR 2023DistilPose: Tokenized Pose Regression With Heatmap Distillation
• NeurIPS 2026DASS: A Solver-Agnostic Dynamic Auxiliary Search Strategy for Symbolic Regression
• ICML 2026AutoNumerics-Zero: Automated Discovery of State-of-the-Art Mathematical Functions
• ICLR 2026GenSR: Symbolic regression based on equation generative space
• CVPR 2026Sparse-View Localization via Online Neural 3D Regression

全部符号回归论文

共 75 篇，第 1 / 2 页

1. CVPR 2025R-SCoRe: Revisiting Scene Coordinate Regression for Robust Large-Scale Visual Localization
2. CVPR 2025Scene-agnostic Pose Regression for Visual Localization
3. ICLR 2025A Graph Enhanced Symbolic Discovery Framework For Efficient Logic Optimization
4. ICLR 2025Approximating Full Conformal Prediction for Neural Network Regression with Gauss-Newton Influence
5. ICLR 2025Attention layers provably solve single-location regression
6. ICLR 2025Bayesian Image Regression with Soft-thresholded Conditional Autoregressive Prior
7. ICLR 2025Benign Overfitting in Out-of-Distribution Generalization of Linear Models
8. ICLR 2025Can Large Language Models Understand Symbolic Graphics Programs?
9. ICLR 2025Discovering Group Structures via Unitary Representation Learning
10. ICLR 2025Distribution-Free Data Uncertainty for Neural Network Regression
11. ICLR 2025Implicit Bias of Mirror Flow for Shallow Neural Networks in Univariate Regression
12. ICLR 2025Improving Deep Regression with Tightness
13. ICLR 2025KinFormer: Generalizable Dynamical Symbolic Regression for Catalytic Organic Reaction Kinetics
14. ICLR 2025LLM-SR: Scientific Equation Discovery via Programming with Large Language Models
15. ICLR 2025Near-optimal Active Regression of Single-Index Models
16. ICLR 2025No Equations Needed: Learning System Dynamics Without Relying on Closed-Form ODEs
17. ICLR 2025ParFam -- (Neural Guided) Symbolic Regression via Continuous Global Optimization
18. ICLR 2025RAG-SR: Retrieval-Augmented Generation for Neural Symbolic Regression
19. ICLR 2025RegMix: Data Mixture as Regression for Language Model Pre-training
20. ICLR 2025State Space Models are Provably Comparable to Transformers in Dynamic Token Selection
21. ICLR 2025Symbolic regression via MDLformer-guided search: from minimizing prediction error to minimizing description length
22. ICLR 2025Towards Learning High-Precision Least Squares Algorithms with Sequence Models
23. ICLR 2025Transformers Handle Endogeneity in In-Context Linear Regression
24. ICLR 2025TRENDy: Temporal Regression of Effective Nonlinear Dynamics
25. ICML 2025A Likelihood Based Approach to Distribution Regression Using Conditional Deep Generative Models
26. ICML 2025Ab Initio Nonparametric Variable Selection for Scalable Symbolic Regression with Large $p$
27. ICML 2025Batch List-Decodable Linear Regression via Higher Moments
28. ICML 2025Benefits of Early Stopping in Gradient Descent for Overparameterized Logistic Regression
29. ICML 2025Curvature Enhanced Data Augmentation for Regression
30. ICML 2025Dataflow-Guided Neuro-Symbolic Language Models for Type Inference
31. ICML 2025Exact Recovery of Sparse Binary Vectors from Generalized Linear Measurements
32. ICML 2025Exploring Invariance in Images through One-way Wave Equations
33. ICML 2025Fast Min-$\epsilon$ Segmented Regression using Constant-Time Segment Merging
34. ICML 2025FEAT-KD: Learning Concise Representations for Single and Multi-Target Regression via TabNet Knowledge Distillation
35. ICML 2025Fully Heteroscedastic Count Regression with Deep Double Poisson Networks
36. ICML 2025Gradient Aligned Regression via Pairwise Losses
37. ICML 2025Heavy-Tailed Linear Bandits: Huber Regression with One-Pass Update
38. ICML 2025Learning Curves of Stochastic Gradient Descent in Kernel Regression
39. ICML 2025LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Models
40. ICML 2025Meta Optimality for Demographic Parity Constrained Regression via Post-Processing
41. ICML 2025On Volume Minimization in Conformal Regression
42. ICML 2025Pareto-Optimal Fronts for Benchmarking Symbolic Regression Algorithms
43. ICML 2025PRIME: Deep Imbalanced Regression with Proxies
44. ICML 2025Sharp Generalization for Nonparametric Regression by Over-Parameterized Neural Networks: A Distribution-Free Analysis in Spherical Covariate
45. ICML 2025Theoretical Limitations of Ensembles in the Age of Overparameterization
46. NeurIPS 2025A Closer Look at NTK Alignment: Linking Phase Transitions in Deep Image Regression
47. NeurIPS 2025A learnability analysis on neuro-symbolic learning
48. NeurIPS 2025Adversarial Robustness of Nonparametric Regression
49. NeurIPS 2025Ascent Fails to Forget
50. NeurIPS 2025Compact Memory for Continual Logistic Regression

常见问题

2023–2026 年四大会议一共有多少篇符号回归论文？

共 316 篇，按年份分别是 2023 年 63 篇、2024 年 81 篇、2025 年 75 篇、2026 年 97 篇。

NeurIPS 2026 有多少篇符号回归论文？

NeurIPS 2026 有 34 篇符号回归论文，上一届是 30 篇。

ICLR 2026 有多少篇符号回归论文？

ICLR 2026 有 24 篇符号回归论文，上一届是 22 篇。

ICML 2026 有多少篇符号回归论文？

ICML 2026 有 36 篇符号回归论文，上一届是 21 篇。

CVPR 2026 有多少篇符号回归论文？

CVPR 2026 有 3 篇符号回归论文，上一届是 2 篇。

哪个会议的符号回归论文最多？

NeurIPS，2023–2026 年共 118 篇。