{"total":91,"rows":[{"id":"2410.07149","title":"Towards Interpreting Visual Information Processing in Vision-Language Models","created":"2024-10-09","cited_by_count":122,"area":"mllm"},{"id":"2504.02821","title":"Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models","created":"2025-04-03","cited_by_count":71,"area":"mllm"},{"id":"2412.05276","title":"Sparse autoencoders reveal selective remapping of visual concepts during adaptation","created":"2024-12-06","cited_by_count":68,"area":"mechinterp"},{"id":"2502.03714","title":"Universal Sparse Autoencoders: Interpretable Cross-Model Concept Alignment","created":"2025-02-06","cited_by_count":62,"area":"mechinterp"},{"id":"2502.20578","title":"Interpreting CLIP with Hierarchical Sparse Autoencoders","created":"2025-02-27","cited_by_count":60,"area":"mechinterp"},{"id":"2412.07992","title":"Concept Bottleneck Large Language Models","created":"2024-12-11","cited_by_count":43,"area":null},{"id":"2504.08729","title":"Steering CLIP's vision transformer with sparse autoencoders","created":"2025-04-11","cited_by_count":38,"area":"mechinterp"},{"id":"2502.17516","title":"A Survey on Mechanistic Interpretability for Multi-Modal Foundation Models","created":"2025-02-22","cited_by_count":34,"area":"mllm"},{"id":"2503.23283","title":"Language Guided Concept Bottleneck Models for Interpretable Continual Learning","created":"2025-03-30","cited_by_count":28,"area":null},{"id":"2502.06755","title":"Interpretable and Testable Vision Features via Sparse Autoencoders","created":"2025-02-10","cited_by_count":25,"area":"mechinterp"},{"id":"2412.11576","title":"DCBM: Data-Efficient Visual Concept Bottleneck Models","created":"2024-12-16","cited_by_count":16,"area":null},{"id":"2503.07050","title":"TIDE : Temporal-Aware Sparse Autoencoders for Interpretable Diffusion Transformers in Image Generation","created":"2025-03-10","cited_by_count":16,"area":"diffusion"},{"id":"2411.06090","title":"Concept Bottleneck Language Models For protein design","created":"2024-11-09","cited_by_count":15,"area":null},{"id":"2504.19774","title":"If Concept Bottlenecks are the Question, are Foundation Models the Answer?","created":"2025-04-28","cited_by_count":15,"area":null},{"id":"2505.18186","title":"Discovering and Steering Interpretable Concepts in Large Generative Music Models","created":"2025-05-18","cited_by_count":13,"area":"mechinterp"},{"id":"2411.00715","title":"B-cosification: Transforming Deep Neural Networks to be Inherently Interpretable","created":"2024-11-01","cited_by_count":12,"area":null},{"id":"2502.20134","title":"Show and Tell: Visually Explainable Deep Neural Nets via Spatially-Aware Concept Bottleneck Models","created":"2025-02-27","cited_by_count":12,"area":null},{"id":"2505.24492","title":"Object Centric Concept Bottlenecks","created":"2025-05-30","cited_by_count":10,"area":null},{"id":"2508.11277","title":"Probing the Representational Power of Sparse Autoencoders in Vision Models","created":"2025-08-15","cited_by_count":10,"area":"diffusion"},{"id":"2509.21379","title":"SAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse Autoencoders","created":"2025-09-23","cited_by_count":10,"area":"diffusion"},{"id":"2602.06218","title":"Cross-Modal Redundancy and the Geometry of Vision-Language Embeddings","created":"2026-02-05","cited_by_count":10,"area":"mllm"},{"id":"2501.04975","title":"V2C-CBM: Building Concept Bottlenecks with Vision-to-Concept Tokenizer","created":"2025-01-09","cited_by_count":9,"area":null},{"id":"2502.09507","title":"When and How Does CLIP Enable Domain and Compositional Generalization?","created":"2025-02-13","cited_by_count":9,"area":"mllm"},{"id":"2501.02922","title":"Label-free Concept Based Multiple Instance Learning for Gigapixel Histopathology","created":"2025-01-06","cited_by_count":8,"area":null},{"id":"2505.20229","title":"From What to How: Attributing CLIP's Latent Components Reveals Unexpected Semantic Reliance","created":"2025-05-26","cited_by_count":6,"area":"mechinterp"},{"id":"2507.23784","title":"SUB: Benchmarking CBM Generalization via Synthetic Attribute Substitutions","created":"2025-07-31","cited_by_count":6,"area":null},{"id":"2512.10805","title":"Interpretable and Steerable Concept Bottleneck Sparse Autoencoders","created":"2025-12-11","cited_by_count":6,"area":"mechinterp"},{"id":"2510.26186","title":"ConceptScope: Characterizing Dataset Bias via Disentangled Visual Concepts","created":"2025-10-30","cited_by_count":5,"area":null},{"id":"2511.22170","title":"Partially Shared Concept Bottleneck Models","created":"2025-11-27","cited_by_count":5,"area":null},{"id":"2412.16247","title":"Towards scientific discovery with dictionary learning: Extracting biological concepts from microscopy foundation models","created":"2024-12-20","cited_by_count":4,"area":null},{"id":"2502.09018","title":"Zero-shot Concept Bottleneck Models","created":"2025-02-13","cited_by_count":4,"area":null},{"id":"2510.03721","title":"Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Models","created":"2025-10-04","cited_by_count":4,"area":"mllm"},{"id":"2603.07343","title":"Learning Concept Bottleneck Models from Mechanistic Explanations","created":"2026-03-07","cited_by_count":4,"area":null},{"id":"2505.16077","title":"Ensembling Sparse Autoencoders","created":"2025-05-21","cited_by_count":3,"area":"mechinterp"},{"id":"2511.16309","title":"Sparse Autoencoders are Topic Models","created":"2025-11-20","cited_by_count":3,"area":null},{"id":"2512.07355","title":"A Geometric Unification of Concept Learning with Concept Cones","created":"2025-12-08","cited_by_count":3,"area":"mechinterp"},{"id":"2602.23947","title":"Hierarchical Concept-based Interpretable Models","created":"2026-02-27","cited_by_count":3,"area":null},{"id":"2603.28211","title":"Explaining CLIP Zero-shot Predictions Through Concepts","created":"2026-03-30","cited_by_count":3,"area":"mllm"},{"id":"2606.06664","title":"Inside the Visual Mind: Neuroscience-Motivated Concept Circuits for Interpreting and Steering Vision Transformers","created":"2026-06-04","cited_by_count":3,"area":"mechinterp"},{"id":"2503.08384","title":"Prototype-Based Multiple Instance Learning for Gigapixel Whole Slide Image Classification","created":"2025-03-11","cited_by_count":2,"area":null},{"id":"2507.15227","title":"Mammo-SAE: Interpreting Breast Cancer Concept Learning with Sparse Autoencoders","created":"2025-07-21","cited_by_count":2,"area":"mechinterp"},{"id":"2509.19191","title":"Reading Images Like Texts: Sequential Image Understanding in Vision-Language Models","created":"2025-09-23","cited_by_count":2,"area":"mllm"},{"id":"2509.21385","title":"Debugging Concept Bottleneck Models through Removal and Retraining","created":"2025-09-23","cited_by_count":2,"area":null},{"id":"2510.26411","title":"MedSAE: Dissecting MedCLIP Representations with Sparse Autoencoders","created":"2025-10-30","cited_by_count":2,"area":"mechinterp"},{"id":"2601.20028","title":"Decomposing multimodal embedding spaces with group-sparse autoencoders","created":"2026-01-27","cited_by_count":2,"area":"mechinterp"},{"id":"2603.05629","title":"Rethinking Concept Bottleneck Models: From Pitfalls to Solutions","created":"2026-03-05","cited_by_count":2,"area":"lr"},{"id":"2603.22593","title":"Language Models Can Explain Visual Features via Steering","created":"2026-03-23","cited_by_count":2,"area":"mechinterp"},{"id":"2604.17089","title":"Tree of Concepts: Interpretable Continual Learners in Non-Stationary Clinical Domains","created":"2026-04-18","cited_by_count":2,"area":null},{"id":"2605.16405","title":"Concepts Worth Having: Refining VLM-Guided Concept Bottleneck Models with Minimal Annotations","created":"2026-05-13","cited_by_count":2,"area":null},{"id":"2606.24716","title":"Evaluating the Interpretability of Sparse Autoencoders with Concept Annotations","created":"2026-06-23","cited_by_count":2,"area":"mllm"},{"id":"2503.10981","title":"CLIP-Free, Label Free, Unsupervised Concept Bottleneck Models","created":"2025-03-14","cited_by_count":1,"area":null},{"id":"2504.12197","title":"Beyond Patches: Mining Interpretable Part-Prototypes for Explainable AI","created":"2025-04-16","cited_by_count":1,"area":null},{"id":"2504.13140","title":"PCBEAR: Pose Concept Bottleneck for Explainable Action Recognition","created":"2025-04-17","cited_by_count":1,"area":null},{"id":"2507.18512","title":"Explaining How Visual, Textual and Multimodal Encoders Share Concepts","created":"2025-07-24","cited_by_count":1,"area":"mechinterp"},{"id":"2509.20376","title":"ConceptViz: A Visual Analytics Approach for Exploring Concepts in Large Language Models","created":"2025-09-20","cited_by_count":1,"area":"mechinterp"},{"id":"2509.20899","title":"Concepts in Motion: Temporal Concept Bottleneck Model for Interpretable Video Classification","created":"2025-09-25","cited_by_count":1,"area":null},{"id":"2510.25512","title":"FaCT: Faithful Concept Traces for Explaining Neural Network Decisions","created":"2025-10-29","cited_by_count":1,"area":null},{"id":"2511.17735","title":"Towards Open-Ended Visual Scientific Discovery with Sparse Autoencoders","created":"2025-11-21","cited_by_count":1,"area":"mechinterp"},{"id":"2602.07311","title":"LUCID-SAE: Learning Unified Vision-Language Sparse Codes for Interpretable Concept Discovery","created":"2026-02-07","cited_by_count":1,"area":"mechinterp"},{"id":"2603.09787","title":"What is Missing? Explaining Neurons Activated by Absent Concepts","created":"2026-03-10","cited_by_count":1,"area":null},{"id":"2604.05724","title":"Beyond Semantics: Disentangling Information Scope in Sparse Autoencoders for CLIP","created":"2026-04-07","cited_by_count":1,"area":"mechinterp"},{"id":"2605.17504","title":"A Distributional View for Visual Mechanistic Interpretability: KL-Minimal Soft-Constraint Principle","created":"2026-05-17","cited_by_count":1,"area":"mechinterp"},{"id":"2606.00082","title":"Hoeffding Concept Bottleneck Models with Applications to Overhead Images","created":"2026-05-22","cited_by_count":1,"area":null},{"id":"2606.07451","title":"TEVI: Text-Conditioned Editing of Visual Representations via Sparse Autoencoders for Improved Vision-Language Alignment","created":"2026-06-05","cited_by_count":1,"area":"mllm"},{"id":"2606.22994","title":"Do Sparse Autoencoders Learn Meaningful Concept Hierarchies?","created":"2026-06-22","cited_by_count":1,"area":"mechinterp"},{"id":"2607.00578","title":"Caption Bottleneck Models","created":"2026-07-01","cited_by_count":1,"area":null},{"id":"2502.05950","title":"Survival Concept-Based Learning Models","created":"2025-02-09","cited_by_count":0,"area":null},{"id":"2509.15272","title":"Which Direction to Choose? An Analysis on the Representation Power of Self-Supervised ViTs in Downstream Tasks","created":"2025-09-18","cited_by_count":0,"area":null},{"id":"2509.22015","title":"Concept-SAE: A Controllable and Invertible Concept Interface for Sparse Autoencoders","created":"2025-09-26","cited_by_count":0,"area":"mechinterp"},{"id":"2510.04180","title":"Spatially Grounded Concept-Based Image Classification","created":"2025-10-05","cited_by_count":0,"area":null},{"id":"2512.03578","title":"When, How Long and How Much? Interpretable Neural Networks for Time Series Regression by Learning to Mask and Aggregate","created":"2025-12-03","cited_by_count":0,"area":null},{"id":"2512.11263","title":"Features Emerge as Discrete States: The First Application of SAEs to 3D Representations","created":"2025-12-12","cited_by_count":0,"area":"mechinterp"},{"id":"2512.19734","title":"The Deleuzian Representation Hypothesis","created":"2025-12-17","cited_by_count":0,"area":"mechinterp"},{"id":"2601.12303","title":"Concepts from Representations: Post-hoc Concept Bottleneck Models via Sparse Decomposition of Visual Representations","created":"2026-01-18","cited_by_count":0,"area":null},{"id":"2602.15650","title":"Concept-Enhanced Multimodal RAG: Towards Interpretable and Accurate Radiology Report Generation","created":"2026-02-17","cited_by_count":0,"area":"mllm"},{"id":"2603.22624","title":"Toward Faithful Segmentation Attribution via Benchmarking and Dual-Evidence Fusion","created":"2026-03-23","cited_by_count":0,"area":null},{"id":"2604.08846","title":"Dictionary-Aligned Concept Control for Safeguarding Multimodal LLMs","created":"2026-04-10","cited_by_count":0,"area":"mllm"},{"id":"2604.11868","title":"MedConcept: Unsupervised Concept Discovery for Interpretability in Medical VLMs","created":"2026-04-13","cited_by_count":0,"area":"mllm"},{"id":"2605.06095","title":"Metonymy in vision models undermines attention-based interpretability","created":"2026-05-07","cited_by_count":0,"area":null},{"id":"2605.06440","title":"Hyperbolic Concept Bottleneck Models","created":"2026-05-07","cited_by_count":0,"area":null},{"id":"2605.14761","title":"AI Outperforms Humans in Personalized Image Aesthetics Assessment via LLM-Based Interviews and Semantic Feature Extraction","created":"2026-05-14","cited_by_count":0,"area":null},{"id":"2605.22679","title":"Conceptualizing Embeddings: Sparse Disentanglement for Vision-Language Models","created":"2026-05-21","cited_by_count":0,"area":"mllm"},{"id":"2606.04326","title":"Measuring What Matters: Synthetic Benchmarks for Concept Bottleneck Models","created":"2026-06-03","cited_by_count":0,"area":null},{"id":"2606.10669","title":"In Defense of Information Leakage in Concept-based Models","created":"2026-06-09","cited_by_count":0,"area":null},{"id":"2606.29888","title":"Same Concept, Different Directions: Cross-Modal Feature Heterogeneity in Sparse Autoencoders","created":"2026-06-29","cited_by_count":0,"area":"mllm"},{"id":"2607.04593","title":"TORINO: Token Reduction via Interpretable Concept Overlap in Vision-Language Models","created":"2026-07-06","cited_by_count":0,"area":"mllm"},{"id":"2608.15404","title":"CBX-Bench: A Human-Aligned MLLM Council for Benchmarking Concept Bottleneck Model Explanations","created":"2026-08-15","cited_by_count":0,"area":"alignment"},{"id":"2608.17063","title":"J-Miner: Recovering the Decision Logic of Fine-Tuned LLM Classifiers as Compact Rules","created":"2026-08-17","cited_by_count":0,"area":null},{"id":"2608.20117","title":"SAE-Xplainers: Rule-Based Feature Interpretation for Extreme Earth Events","created":"2026-08-20","cited_by_count":0,"area":"mechinterp"},{"id":"2609.37786","title":"CHOQOLATE: Organizing Concept Bottleneck Latent Spaces with Choquet Integrals","created":"2026-09-29","cited_by_count":0,"area":"mllm"},{"id":"2609.39625","title":"D-Scope: Decomposing and Steering Diffusion Transformers with Sparse Autoencoders","created":"2026-09-30","cited_by_count":0,"area":"diffusion"}],"next_cursor":null}