概念解释论文趋势 2023–2026
TCAV 等
统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 342 篇被归入该主题的论文。
主要结论
- 概念解释论文从 2023 年的 91 篇增长到 2026 年的 117 篇，三年间扩大到约 1.3 倍。
- NeurIPS 是发概念解释论文最多的会议，2023–2026 年共 134 篇，占四个会议总量的 39%。
- 交叉方向里涨得最快的是多模态大模型：同时属于两个主题的论文从 2023 年的 4 篇增至 2026 年的 23 篇。
- 在「可解释性」的 5 个主题里，它的 2026 年论文数排第 3；同组增长最快的机制解释是 8.1 倍，概念解释是 1.3 倍。
各会议论文数
按年份和会议统计
| 年份 | NeurIPS | ICLR | ICML | CVPR | 合计 |
|---|---|---|---|---|---|
| 2023 | 48 | 9 | 27 | 7 | 91 |
| 2024 | 21 | 9 | 20 | 13 | 63 |
| 2025 | 25 | 19 | 11 | 16 | 71 |
| 2026 | 40 | 24 | 20 | 33 | 117 |
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 2026CENDRe: Concept Extraction with Natural Domain Representations
- ICML 2026Evidential Copula Concept Embedding Models
- ICLR 2026Decoupling the Class Label and the Target Concept in Machine Unlearning
- CVPR 2026Agile Deliberation Concept Deliberation for Subjective Visual Classification
- NeurIPS 2025An Analysis of Concept Bottleneck Models: Measuring, Understanding, and Mitigating the Impact of Noisy Annotations
- ICML 2025Concept-Centric Token Interpretation for Vector-Quantized Generative Models
- ICLR 2025Causal Concept Graph Models: Beyond Causal Opacity in Deep Learning
- CVPR 2025Discovering Fine-Grained Visual-Concept Relations by Disentangled Optimal Transport Concept Bottleneck Models
- NeurIPS 2024[Re] On the Reproducibility of Post-Hoc Concept Bottleneck Models
- ICML 2024ContPhy: Continuum Physical Concept Learning and Reasoning from Videos
- ICLR 2024AttEXplore: Attribution for Explanation with model parameters eXploration
- CVPR 2024Fine-Grained Bipartite Concept Factorization for Clustering
- NeurIPS 2023[Re] Hierarchical Shrinkage: Improving the Accuracy and Interpretability of Tree-Based Methods
- ICML 2023A Closer Look at the Intervention Procedure of Concept Bottleneck Models
- ICLR 2023Concept Gradient: Concept-based Interpretation Without Linear Assumption
- CVPR 2023CRAFT: Concept Recursive Activation FacTorization for Explainability
- NeurIPS 2026Concept frustration: Aligning human concepts and machine representations
- ICML 2026On the Variability of Concept Activation Vectors
- ICLR 2026Hierarchical Concept-based Interpretable Models
- CVPR 2026DCoAR Deep Concept Injection into Unified Autoregressive Models for Personalized
全部概念解释论文
共 342 篇，第 6 / 7 页
- NeurIPS 2024Uncovering Safety Risks of Large Language Models through Concept Activation Vector
- CVPR 20233D Concept Learning and Reasoning From Multi-View Images
- CVPR 2023CRAFT: Concept Recursive Activation FacTorization for Explainability
- CVPR 2023Grounding Counterfactual Explanation of Image Classifiers to Textual Concept Space
- CVPR 2023IDGI: A Framework To Eliminate Explanation Noise From Integrated Gradients
- CVPR 2023Overlooked Factors in Concept-Based Explanations: Dataset Choice, Concept Learnability, and Human Capability
- CVPR 2023S3C: Semi-Supervised VQA Natural Language Explanation via Self-Critical Learning
- CVPR 2023Spatial-Temporal Concept Based Explanation of 3D ConvNets
- ICLR 2023A Minimalist Dataset for Systematic Generalization of Perception, Syntax, and Semantics
- ICLR 2023Concept Gradient: Concept-based Interpretation Without Linear Assumption
- ICLR 2023Concept-level Debugging of Part-Prototype Networks
- ICLR 2023Global Explainability of GNNs via Logic Combination of Learned Concepts
- ICLR 2023GNNInterpreter: A Probabilistic Generative Model-Level Explanation for Graph Neural Networks
- ICLR 2023Label-free Concept Bottleneck Models
- ICLR 2023Neural Systematic Binder
- ICLR 2023Post-hoc Concept Bottleneck Models
- ICLR 2023Robust Explanation Constraints for Neural Networks
- ICML 2023A Category-theoretical Meta-analysis of Definitions of Disentanglement
- ICML 2023A Closer Look at the Intervention Procedure of Concept Bottleneck Models
- ICML 2023A Kernel-Based View of Language Model Fine-Tuning
- ICML 2023A Scalable Frank-Wolfe-Based Algorithm for the Max-Cut SDP
- ICML 2023Adaptively Weighted Data Augmentation Consistency Regularization for Robust Optimization under Concept Shift
- ICML 2023Causal Proxy Models for Concept-based Model Explanations
- ICML 2023Cluster Explanation via Polyhedral Descriptions
- ICML 2023Concept-based Explanations for Out-of-Distribution Detectors
- ICML 2023Denoising MCMC for Accelerating Diffusion-Based Generative Models
- ICML 2023Difference-in-Differences Meets Tree-based Methods: Heterogeneous Treatment Effects Estimation with Unmeasured Confounding
- ICML 2023Discover and Cure: Concept-aware Mitigation of Spurious Correlation
- ICML 2023Facial Expression Recognition with Adaptive Frame Rate based on Multiple Testing Correction
- ICML 2023Flash: Concept Drift Adaptation in Federated Learning
- ICML 2023Geometric Clifford Algebra Networks
- ICML 2023Interpretable Neural-Symbolic Concept Reasoning
- ICML 2023Lowering the Pre-training Tax for Gradient-based Subset Training: A Lightweight Distributed Pre-Training Toolkit
- ICML 2023Neuro-Symbolic Continual Learning: Knowledge, Reasoning Shortcuts and Concept Rehearsal
- ICML 2023On the Relationship Between Explanation and Prediction: A Causal View
- ICML 2023Probabilistic Concept Bottleneck Models
- ICML 2023PWSHAP: A Path-Wise Explanation Model for Targeted Variables
- ICML 2023Refining Generative Process with Discriminator Guidance in Score-based Diffusion Models
- ICML 2023Robust Explanation for Free or At the Cost of Faithfulness
- ICML 2023SlotGAT: Slot-based Message Passing for Heterogeneous Graphs
- ICML 2023The Implicit Regularization of Dynamical Stability in Stochastic Gradient Descent
- ICML 2023The Virtues of Laziness in Model-based RL: A Unified Objective and Algorithms
- ICML 2023Towards Bridging the Gaps between the Right to Explanation and the Right to be Forgotten
- ICML 2023Towards Trustworthy Explanation: On Causal Rationalization
- NeurIPS 2023[Re] Hierarchical Shrinkage: Improving the Accuracy and Interpretability of Tree-Based Methods
- NeurIPS 2023$k$-Means Clustering with Distance-Based Privacy
- NeurIPS 2023A Dynamical System View of Langevin-Based Non-Convex Sampling
- NeurIPS 2023A Holistic Approach to Unifying Automatic Concept Extraction and Concept Importance Estimation
- NeurIPS 2023A Theory of Transfer-Based Black-Box Attacks: Explanation and Implications
- NeurIPS 2023A*Net: A Scalable Path-based Reasoning Approach for Knowledge Graphs
在 Claude、Cursor 里直接问这个趋势
接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。
可以直接复制这句话问 AI
用 paperlist 查概念解释在 NeurIPS 2023–2026 的论文趋势，哪一年涨得最快？
AI 会调用的工具：get_topic_stats {"topic": "concept-based-explanation"}
常见问题
- 2023–2026 年四大会议一共有多少篇概念解释论文？
- 共 342 篇，按年份分别是 2023 年 91 篇、2024 年 63 篇、2025 年 71 篇、2026 年 117 篇。
- NeurIPS 2026 有多少篇概念解释论文？
- NeurIPS 2026 有 40 篇概念解释论文，上一届是 25 篇。
- ICLR 2026 有多少篇概念解释论文？
- ICLR 2026 有 24 篇概念解释论文，上一届是 19 篇。
- ICML 2026 有多少篇概念解释论文？
- ICML 2026 有 20 篇概念解释论文，上一届是 11 篇。
- CVPR 2026 有多少篇概念解释论文？
- CVPR 2026 有 33 篇概念解释论文，上一届是 16 篇。
- 哪个会议的概念解释论文最多？
- NeurIPS，2023–2026 年共 134 篇。