{"total":62,"rows":[{"id":"2502.12892","title":"Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models","created":"2025-02-18","cited_by_count":73,"area":"mechinterp"},{"id":"2504.02821","title":"Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models","created":"2025-04-03","cited_by_count":71,"area":"mllm"},{"id":"2503.01822","title":"Projecting Assumptions: The Duality Between Sparse Autoencoders and Concept Geometry","created":"2025-03-03","cited_by_count":61,"area":"mechinterp"},{"id":"2504.02922","title":"Overcoming Sparsity Artifacts in Crosscoders to Interpret Chat-Tuning","created":"2025-04-03","cited_by_count":34,"area":null},{"id":"2504.11695","title":"Interpreting the linear structure of vision-language model embedding spaces","created":"2025-04-16","cited_by_count":32,"area":"mllm"},{"id":"2502.06755","title":"Interpretable and Testable Vision Features via Sparse Autoencoders","created":"2025-02-10","cited_by_count":25,"area":"mechinterp"},{"id":"2503.01824","title":"From superposition to sparse codes: interpretable representations in neural networks","created":"2025-03-03","cited_by_count":16,"area":null},{"id":"2506.15963","title":"On the Limits of Sparse Autoencoders: A Theoretical Framework and Reweighted Remedy","created":"2025-06-19","cited_by_count":14,"area":"mechinterp"},{"id":"2510.00404","title":"AbsTopK: Rethinking Sparse Autoencoders For Bidirectional Features","created":"2025-10-01","cited_by_count":9,"area":"lr"},{"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":"2606.25234","title":"Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds","created":"2026-06-23","cited_by_count":5,"area":null},{"id":"2603.07343","title":"Learning Concept Bottleneck Models from Mechanistic Explanations","created":"2026-03-07","cited_by_count":4,"area":null},{"id":"2504.06486","title":"Mind the (Data) Gap: Evaluating Vision Systems in Small Data Applications","created":"2025-04-08","cited_by_count":3,"area":null},{"id":"2507.16069","title":"Interpreting CFD Surrogates through Sparse Autoencoders","created":"2025-07-21","cited_by_count":3,"area":"mechinterp"},{"id":"2510.27413","title":"Atlas-Alignment: Making Interpretability Transferable Across Language Models","created":"2025-10-31","cited_by_count":3,"area":"mechinterp"},{"id":"2602.00621","title":"Towards Interpretable Hallucination Analysis and Mitigation in LVLMs via Contrastive Neuron Steering","created":"2026-01-31","cited_by_count":3,"area":"mechinterp"},{"id":"2605.10971","title":"Steering Without Breaking: Mechanistically Informed Interventions for Discrete Diffusion Language Models","created":"2026-05-08","cited_by_count":3,"area":"diffusion"},{"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":"2506.01247","title":"Beyond Interpretability: When, Why, and How Sparse Autoencoders Enable Label-Free Visual Steering","created":"2025-06-02","cited_by_count":2,"area":"mechinterp"},{"id":"2508.03649","title":"Cross-Model Semantics in Representation Learning","created":"2025-08-05","cited_by_count":2,"area":null},{"id":"2510.14053","title":"Position: Require Frontier AI Labs To Release Small \"Analog\" Models","created":"2025-10-15","cited_by_count":2,"area":null},{"id":"2601.20028","title":"Decomposing multimodal embedding spaces with group-sparse autoencoders","created":"2026-01-27","cited_by_count":2,"area":"mechinterp"},{"id":"2605.13675","title":"Characterizing Universal Object Representations Across Vision Models","created":"2026-05-13","cited_by_count":2,"area":null},{"id":"2608.09521","title":"One Adapter Pair per Model: A Universal Activation Interface for Language Models","created":"2026-08-10","cited_by_count":2,"area":"mechinterp"},{"id":"2507.18512","title":"Explaining How Visual, Textual and Multimodal Encoders Share Concepts","created":"2025-07-24","cited_by_count":1,"area":"mechinterp"},{"id":"2508.07833","title":"MIMIC: Multimodal Inversion for Model Interpretation and Conceptualization","created":"2025-08-11","cited_by_count":1,"area":"mllm"},{"id":"2512.07245","title":"Zero-Shot Textual Explanations via Translating Decision-Critical Features","created":"2025-12-08","cited_by_count":1,"area":null},{"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":"2604.04946","title":"Sparse Autoencoders as a Steering Basis for Phase Synchronization in Graph-Based CFD Surrogates","created":"2026-03-28","cited_by_count":1,"area":"mechinterp"},{"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":"2605.25941","title":"Where Concept Erasure Should Occur: Concept-Layer Alignment in Text-to-Video Diffusion Models","created":"2026-05-25","cited_by_count":1,"area":"diffusion"},{"id":"2606.09653","title":"A Unifying Framework for Concept-Based Representational Similarity","created":"2026-06-08","cited_by_count":1,"area":null},{"id":"2607.08605","title":"When Structured Sparse Autoencoders Learn Consistent Concepts Across Modalities","created":"2026-07-09","cited_by_count":1,"area":"mechinterp"},{"id":"2505.23917","title":"Representational Difference Explanations","created":"2025-05-29","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":"2604.13304","title":"Can Cross-Layer Transcoders Replace Vision Transformer Activations? An Interpretable Perspective on Vision","created":"2026-04-14","cited_by_count":0,"area":"mechinterp"},{"id":"2605.20337","title":"Capability $\\neq$ Interpretability: Human Interpretability of Vision Foundation Models","created":"2026-05-19","cited_by_count":0,"area":null},{"id":"2605.23040","title":"Steered Generation via Gradient-Based Optimization on Sparse Query Features","created":"2026-05-21","cited_by_count":0,"area":"mechinterp"},{"id":"2605.26921","title":"Similarity-based representation factorization for revealing interpretable dimensions in representational data","created":"2026-05-26","cited_by_count":0,"area":null},{"id":"2605.30049","title":"Robust and Generalizable Safety Steering for Text-to-Image Diffusion Transformers","created":"2026-05-28","cited_by_count":0,"area":"diffusion"},{"id":"2606.07882","title":"The Cross-Architecture Substrate: A Domain-Transcendent, Calibration-Surviving Geometric Invariant of Modern Vision Encoders","created":"2026-06-05","cited_by_count":0,"area":null},{"id":"2606.16897","title":"Contrastive-Difference CKA Reveals Concept-Specific Structural Alignment Across Language Model Architectures","created":"2026-06-15","cited_by_count":0,"area":null},{"id":"2607.08499","title":"Cross-seed explainability using Procrustes-conditioned Joint End-to-end Top-K Sparse Autoencoders","created":"2026-07-09","cited_by_count":0,"area":"mechinterp"},{"id":"2607.12094","title":"Sparse Autoencoders for Interpretable Out-of-Distribution Detection","created":"2026-07-13","cited_by_count":0,"area":"mechinterp"},{"id":"2608.00561","title":"Through the LENS: Local Geometric Decomposition of Vision-Language Model Representations","created":"2026-08-01","cited_by_count":0,"area":"mllm"},{"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":"2608.24142","title":"What Does Prompt Learning Change? -A Natural-Language Concept Analysis of Vision-Language Models","created":"2026-08-25","cited_by_count":0,"area":"mllm"},{"id":"2609.04276","title":"FailSAE: Towards Interpretable Failure Prediction for Vision-Language Models via Sparse Autoencoders","created":"2026-09-02","cited_by_count":0,"area":"mllm"},{"id":"2609.04344","title":"SharedSAE: One Feature Dictionary Across Language Models","created":"2026-09-03","cited_by_count":0,"area":"mechinterp"},{"id":"2609.04721","title":"Locating and Steering Refusal Beyond Attention","created":"2026-09-04","cited_by_count":0,"area":null},{"id":"2609.05399","title":"From Interpretability Methods to Interpretable Models","created":"2026-09-04","cited_by_count":0,"area":null},{"id":"2609.08879","title":"Medical AI Encodes a \"Feeling of Error\": Verifying Cancer Segmentation via Internal Concepts","created":"2026-09-08","cited_by_count":0,"area":"mechinterp"},{"id":"2609.13198","title":"SPICE: Simple Polysemantic Feature Interpretation via Clustering-based Explanation","created":"2026-08-14","cited_by_count":0,"area":null},{"id":"2609.29551","title":"HiPACE: Hierarchical Phase-Boundary Analysis and Controlled Evaluation of Feature Absorption in Sparse Autoencoders","created":"2026-08-26","cited_by_count":0,"area":"mechinterp"},{"id":"2609.31996","title":"Can Circuit Alignment Predict OOD Generalization?","created":"2026-09-25","cited_by_count":0,"area":null},{"id":"2609.35020","title":"Verifying the Linear Representation Hypothesis: How Interpretable Are Vision SAEs?","created":"2026-09-28","cited_by_count":0,"area":"mechinterp"},{"id":"2610.00848","title":"Geometric Similarity in VLM Low-Level Vision Representations","created":"2026-10-01","cited_by_count":0,"area":"mllm"},{"id":"2610.01864","title":"From Isolated Feature to Orbits: Discovering Music Concepts via Multi-SAE Alignment","created":"2026-10-01","cited_by_count":0,"area":"mechinterp"}],"next_cursor":null}