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 695 篇，第 14 / 14 页

651. ICML 2023Defects of Convolutional Decoder Networks in Frequency Representation
652. ICML 2023DRCFS: Doubly Robust Causal Feature Selection
653. ICML 2023Faith-Shap: The Faithful Shapley Interaction Index
654. ICML 2023Feature Expansion for Graph Neural Networks
655. ICML 2023Feature learning in deep classifiers through Intermediate Neural Collapse
656. ICML 2023Feature Programming for Multivariate Time Series Prediction
657. ICML 2023Global optimality of Elman-type RNNs in the mean-field regime
658. ICML 2023GRAFENNE: Learning on Graphs with Heterogeneous and Dynamic Feature Sets
659. ICML 2023Graph Positional Encoding via Random Feature Propagation
660. ICML 2023Hybrid Energy Based Model in the Feature Space for Out-of-Distribution Detection
661. ICML 2023Learning Perturbations to Explain Time Series Predictions
662. ICML 2023Learning to Maximize Mutual Information for Dynamic Feature Selection
663. ICML 2023LESS-VFL: Communication-Efficient Feature Selection for Vertical Federated Learning
664. ICML 2023MetaModulation: Learning Variational Feature Hierarchies for Few-Shot Learning with Fewer Tasks
665. ICML 2023Monge, Bregman and Occam: Interpretable Optimal Transport in High-Dimensions with Feature-Sparse Maps
666. ICML 2023Simple and Fast Group Robustness by Automatic Feature Reweighting
667. ICML 2023Sketched Ridgeless Linear Regression: The Role of Downsampling
668. ICML 2023SRATTA: Sample Re-ATTribution Attack of Secure Aggregation in Federated Learning.
669. ICML 2023The Benefits of Mixup for Feature Learning
670. ICML 2023Towards credible visual model interpretation with path attribution
671. NeurIPS 2023$\mathcal{M}^4$: A Unified XAI Benchmark for Faithfulness Evaluation of Feature Attribution Methods across Metrics, Modalities and Models
672. NeurIPS 2023A Neural Collapse Perspective on Feature Evolution in Graph Neural Networks
673. NeurIPS 2023A Performance-Driven Benchmark for Feature Selection in Tabular Deep Learning
674. NeurIPS 2023Adversarial Learning for Feature Shift Detection and Correction
675. NeurIPS 2023Better Private Linear Regression Through Better Private Feature Selection
676. NeurIPS 2023Discriminative Feature Attributions: Bridging Post Hoc Explainability and Inherent Interpretability
677. NeurIPS 2023Error Bounds for Learning with Vector-Valued Random Features
678. NeurIPS 2023Feature Adaptation for Sparse Linear Regression
679. NeurIPS 2023Feature learning via mean-field Langevin dynamics: classifying sparse parities and beyond
680. NeurIPS 2023Feature Likelihood Divergence: Evaluating the Generalization of Generative Models Using Samples
681. NeurIPS 2023FedFed: Feature Distillation against Data Heterogeneity in Federated Learning
682. NeurIPS 2023GAIA: Delving into Gradient-based Attribution Abnormality for Out-of-distribution Detection
683. NeurIPS 2023Going Beyond Linear Mode Connectivity: The Layerwise Linear Feature Connectivity
684. NeurIPS 2023Joint Feature and Differentiable $ k $-NN Graph Learning using Dirichlet Energy
685. NeurIPS 2023Learning Curves for Deep Structured Gaussian Feature Models
686. NeurIPS 2023Learning Curves for Noisy Heterogeneous Feature-Subsampled Ridge Ensembles
687. NeurIPS 2023Model and Feature Diversity for Bayesian Neural Networks in Mutual Learning
688. NeurIPS 2023Reading Relevant Feature from Global Representation Memory for Visual Object Tracking
689. NeurIPS 2023SHAP-IQ: Unified Approximation of any-order Shapley Interactions
690. NeurIPS 2023SmoothHess: ReLU Network Feature Interactions via Stein's Lemma
691. NeurIPS 2023Stability Guarantees for Feature Attributions with Multiplicative Smoothing
692. NeurIPS 2023TexQ: Zero-shot Network Quantization with Texture Feature Distribution Calibration
693. NeurIPS 2023Towards Efficient Pre-Trained Language Model via Feature Correlation Distillation
694. NeurIPS 2023Towards Hybrid-grained Feature Interaction Selection for Deep Sparse Network
695. NeurIPS 2023Towards Stable Backdoor Purification through Feature Shift Tuning

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。