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 43 篇，第 1 / 1 页

1. CVPR 2023A Practical Upper Bound for the Worst-Case Attribution Deviations
2. CVPR 2023Adaptive Assignment for Geometry Aware Local Feature Matching
3. CVPR 2023Adaptive Spot-Guided Transformer for Consistent Local Feature Matching
4. CVPR 2023AdaptiveMix: Improving GAN Training via Feature Space Shrinkage
5. CVPR 2023AGAIN: Adversarial Training With Attribution Span Enlargement and Hybrid Feature Fusion
6. CVPR 2023Block Selection Method for Using Feature Norm in Out-of-Distribution Detection
7. CVPR 2023CafeBoost: Causal Feature Boost To Eliminate Task-Induced Bias for Class Incremental Learning
8. CVPR 2023Camouflaged Object Detection With Feature Decomposition and Edge Reconstruction
9. CVPR 2023Clothing-Change Feature Augmentation for Person Re-Identification
10. CVPR 2023Correspondence Transformers With Asymmetric Feature Learning and Matching Flow Super-Resolution
11. CVPR 2023Curvature-Balanced Feature Manifold Learning for Long-Tailed Classification
12. CVPR 2023Data-Driven Feature Tracking for Event Cameras
13. CVPR 2023Data-Free Knowledge Distillation via Feature Exchange and Activation Region Constraint
14. CVPR 2023Discriminative Co-Saliency and Background Mining Transformer for Co-Salient Object Detection
15. CVPR 2023Discriminator-Cooperated Feature Map Distillation for GAN Compression
16. CVPR 2023DKM: Dense Kernelized Feature Matching for Geometry Estimation
17. CVPR 2023Efficient View Synthesis and 3D-Based Multi-Frame Denoising With Multiplane Feature Representations
18. CVPR 2023FAC: 3D Representation Learning via Foreground Aware Feature Contrast
19. CVPR 2023FCC: Feature Clusters Compression for Long-Tailed Visual Recognition
20. CVPR 2023Feature Separation and Recalibration for Adversarial Robustness
21. CVPR 2023Feature Shrinkage Pyramid for Camouflaged Object Detection With Transformers
22. CVPR 2023FeatureBooster: Boosting Feature Descriptors With a Lightweight Neural Network
23. CVPR 2023Fix the Noise: Disentangling Source Feature for Controllable Domain Translation
24. CVPR 2023GCFAgg: Global and Cross-View Feature Aggregation for Multi-View Clustering
25. CVPR 2023Generalizable Local Feature Pre-Training for Deformable Shape Analysis
26. CVPR 2023Gradient-Based Uncertainty Attribution for Explainable Bayesian Deep Learning
27. CVPR 2023Harmonious Feature Learning for Interactive Hand-Object Pose Estimation
28. CVPR 2023Initialization Noise in Image Gradients and Saliency Maps
29. CVPR 2023Introducing Competition To Boost the Transferability of Targeted Adversarial Examples Through Clean Feature Mixup
30. CVPR 2023MetaFusion: Infrared and Visible Image Fusion via Meta-Feature Embedding From Object Detection
31. CVPR 2023Multi-Modal Gait Recognition via Effective Spatial-Temporal Feature Fusion
32. CVPR 2023Multi-Modal Learning With Missing Modality via Shared-Specific Feature Modelling
33. CVPR 2023NeRFLight: Fast and Light Neural Radiance Fields Using a Shared Feature Grid
34. CVPR 2023Panoptic Compositional Feature Field for Editable Scene Rendering With Network-Inferred Labels via Metric Learning
35. CVPR 2023PATS: Patch Area Transportation With Subdivision for Local Feature Matching
36. CVPR 2023Preserving Linear Separability in Continual Learning by Backward Feature Projection
37. CVPR 2023Progressive Open Space Expansion for Open-Set Model Attribution
38. CVPR 2023Rethinking Feature-Based Knowledge Distillation for Face Recognition
39. CVPR 2023Rethinking Gradient Projection Continual Learning: Stability / Plasticity Feature Space Decoupling
40. CVPR 2023SCConv: Spatial and Channel Reconstruction Convolution for Feature Redundancy
41. CVPR 2023SFD2: Semantic-Guided Feature Detection and Description
42. CVPR 2023Shape-Erased Feature Learning for Visible-Infrared Person Re-Identification
43. CVPR 2023UTM: A Unified Multiple Object Tracking Model With Identity-Aware Feature Enhancement

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。