特征归因论文趋势 2023–2026

主要结论

• 特征归因论文从 2023 年的 118 篇增长到 2026 年的 219 篇，三年间扩大到约 1.9 倍。
• NeurIPS 是发特征归因论文最多的会议，2023–2026 年共 193 篇，占四个会议总量的 28%。
• 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 14 篇。
• 在「可解释性」的 5 个主题里，它的 2026 年论文数排第 1；同组增长最快的机制解释是 8.1 倍，特征归因是 1.9 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 2026ADAPT: Hybrid Prompt Optimization for LLM Feature Visualization
• ICML 2026Activation with Intrinsic-Extrinsic Consensus
• ICLR 2026AnyUp: Universal Feature Upsampling
• CVPR 2026A Training-Free Style-Personalization via SVD-Based Feature Decomposition
• NeurIPS 2025Alternating Gradient Flows: A Theory of Feature Learning in Two-layer Neural Networks
• ICML 2025BCE vs. CE in Deep Feature Learning
• ICLR 2025A Differentiable Rank-Based Objective for Better Feature Learning
• CVPR 2025A Simple Data Augmentation for Feature Distribution Skewed Federated Learning
• NeurIPS 2024$\texttt{dattri}$: A Library for Efficient Data Attribution
• ICML 2024Asymptotics of feature learning in two-layer networks after one gradient-step
• ICLR 2024An Intuitive Multi-Frequency Feature Representation for SO(3)-Equivariant Networks
• CVPR 2024DYSON: Dynamic Feature Space Self-Organization for Online Task-Free Class Incremental Learning
• NeurIPS 2023$\mathcal{M}^4$: A Unified XAI Benchmark for Faithfulness Evaluation of Feature Attribution Methods across Metrics, Modalities and Models
• ICML 2023Decoding Layer Saliency in Language Transformers
• ICLR 2023Can We Faithfully Represent Absence States to Compute Shapley Values on a DNN?
• CVPR 2023A Practical Upper Bound for the Worst-Case Attribution Deviations
• NeurIPS 2026Adaptive Feature Propagation for Attribute-Missing Graph Clustering
• ICML 2026Aggregate Models, Not Explanations: Improving Feature Importance Estimation
• ICLR 2026Attribution-Guided Decoding
• CVPR 2026FBTA Enabling Single-GPU End-to-End Gigapixel WSI Classification with Feature Bridging

全部特征归因论文

共 193 篇，第 1 / 4 页

1. NeurIPS 2026ADA: Resolving Attribution Ambiguity in End-to-End Power System Dispatch via Two-Time-Scale Stochastic Approximation
2. NeurIPS 2026ADAPT: Hybrid Prompt Optimization for LLM Feature Visualization
3. NeurIPS 2026Adaptive Feature Propagation for Attribute-Missing Graph Clustering
4. NeurIPS 2026Attribution-Guided Exit Policy for Reliable Early-Exit Inference
5. NeurIPS 2026Avoiding Feature Collapse in Graph ODEs via Hysteretic Topology Evolution
6. NeurIPS 2026Boundary Mass in Feature Space as a Label-Budget Diagnostic for Representations
7. NeurIPS 2026Bug or Feature$^2$: Weight Drift, Activation Sparsity, and Spikes
8. NeurIPS 2026Calibrating Generative Models to Feature Distributions with MMD Finetuning
9. NeurIPS 2026Characterizing the Generalization Error of Random Feature Regression with Arbitrary Data-Augmentation
10. NeurIPS 2026Decomposing Conformal Uncertainty: Calibration- and Instance-Driven Feature Attribution
11. NeurIPS 2026Designing Kernel Surrogate Models for Multimodal Attribution
12. NeurIPS 2026DifFRACT: Diffusion Feature Reconstruction and Attribution for Circuit Tracing
13. NeurIPS 2026Disambiguating 2D-3D Correspondences in Gaussian Splatting-based Feature Fields for Visual Localization
14. NeurIPS 2026EDITORS Know Your Style! Editing LoRA Subspaces for Stylistic Attribution and Imitation
15. NeurIPS 2026Efficient Benchmarking Is Just Feature Selection and Multiple Regression
16. NeurIPS 2026ESENSC: A Polynomial-Time Axiomatic Alternative to SHAP
17. NeurIPS 2026Feature Information Dynamics in Diffusion
18. NeurIPS 2026Federated Logic Gate Networks via Boolean Feature Selection
19. NeurIPS 2026FiLoRA: Focus-and-Ignore LoRA for Controllable Feature Reliance
20. NeurIPS 2026FM-ChangeNet: Learning Change through Pathwise Feature Transport
21. NeurIPS 2026Forward Shapley Scoring for Non-Myopic Active Feature Acquisition
22. NeurIPS 2026FRInGe: Distribution-Space Integrated Gradients with Fisher–Rao Geometry
23. NeurIPS 2026From Baselines to Transport Geodesics: Axiomatic Attribution via Optimal Generative Flows
24. NeurIPS 2026From Isolated feature to Orbits:\\Discovering Music Concepts via Multi-SAE Alignment
25. NeurIPS 2026GraphUAT: Uncertainty Attribution in Graph Neural Networks
26. NeurIPS 2026haphazard: A unified library and benchmark for online learning under varying feature availability
27. NeurIPS 2026How does feature learning change the function space evolution?
28. NeurIPS 2026Hypernetworks for Dynamic Feature Selection
29. NeurIPS 2026Imperfect Influence, Reliable Rankings: A Theory of TRAK for Data Attribution
30. NeurIPS 2026In-Context Benign Overfitting: A Feature-Selection Model in Linear Regression ICL
31. NeurIPS 2026Interpretability-by-Design with Accurate Locally Additive Models and Conditional Feature Effects
32. NeurIPS 2026Intrinsically Interpretable Attention via Sparse Post-Training
33. NeurIPS 2026Invariant Features in Language Models: Geometric Characterization and Model Attribution
34. NeurIPS 2026Is Complex Training Necessary for Long-Tailed OOD Detection? A Re-think from Feature Geometry
35. NeurIPS 2026LDD-RFM: Learnable Domain Decomposition for Random Feature Models via Variable Projection
36. NeurIPS 2026Multi-Objective Reinforcement Learning Using Routed Ensembles and Trajectory Attribution
37. NeurIPS 2026Panoptic Saliency Ranking
38. NeurIPS 2026PE-SHAP: Causally Interpretable Path-Wise Shapley Explanations
39. NeurIPS 2026quanda: An Interpretability Toolkit for Training Data Attribution Evaluation
40. NeurIPS 2026Reading Attribution from Attention: Evidence Heads as Latent Attribution Mechanisms in LLMs
41. NeurIPS 2026Relational Feature Distillation for Lightweight 3D Point Cloud Segmentation
42. NeurIPS 2026Residual Calibration via Local Feature-Space Refinement
43. NeurIPS 2026Reviving the Discarded High-Resolution Feature for Transformer-Based Tiny Object Detection
44. NeurIPS 2026Same Concept, Different Directions: Cross-Modal Feature Heterogeneity in Sparse Autoencoders
45. NeurIPS 2026ScopeSAE: Model-Scope Feature Discovery with Interpretable Layer Selection
46. NeurIPS 2026Semantic Optimal Transport for Sparse Autoencoder Feature Matching and Circuit Compression
47. NeurIPS 2026SGD Provably Prioritizes a Shortcut Spurious Feature in the XOR Model
48. NeurIPS 2026Sharp feature-learning transitions and Bayes-optimal neural scaling laws in extensive-width networks
49. NeurIPS 2026Spectral Insights from the Unconstrained Feature Model for Neural Multi-Output Regression
50. NeurIPS 2026Strategic Feature Selection and Regularization

常见问题

2023–2026 年四大会议一共有多少篇特征归因论文？

共 695 篇，按年份分别是 2023 年 118 篇、2024 年 168 篇、2025 年 190 篇、2026 年 219 篇。

NeurIPS 2026 有多少篇特征归因论文？

NeurIPS 2026 有 61 篇特征归因论文，上一届是 57 篇。

ICLR 2026 有多少篇特征归因论文？

ICLR 2026 有 53 篇特征归因论文，上一届是 42 篇。

ICML 2026 有多少篇特征归因论文？

ICML 2026 有 56 篇特征归因论文，上一届是 51 篇。

CVPR 2026 有多少篇特征归因论文？

CVPR 2026 有 49 篇特征归因论文，上一届是 40 篇。

哪个会议的特征归因论文最多？

NeurIPS，2023–2026 年共 193 篇。