{"total":80,"rows":[{"id":"2504.02821","title":"Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models","created":"2025-04-03","cited_by_count":71,"area":"mllm"},{"id":"2504.12782","title":"Set You Straight: Auto-Steering Denoising Trajectories to Sidestep Unwanted Concepts","created":"2025-04-17","cited_by_count":62,"area":null},{"id":"2503.18878","title":"I Have Covered All the Bases Here: Interpreting Reasoning Features in Large Language Models via Sparse Autoencoders","created":"2025-03-24","cited_by_count":45,"area":"lr"},{"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":"2504.15473","title":"Emergence and Evolution of Interpretable Concepts in Diffusion Models","created":"2025-04-21","cited_by_count":22,"area":"diffusion"},{"id":"2506.23845","title":"Position: Use Sparse Autoencoders to Discover Unknowns","created":"2025-06-30","cited_by_count":19,"area":"mechinterp"},{"id":"2503.14530","title":"SAUCE: Selective Concept Unlearning in Vision-Language Models with Sparse Autoencoders","created":"2025-03-16","cited_by_count":13,"area":"mllm"},{"id":"2508.12535","title":"CorrSteer: Generation-Time LLM Steering via Correlated Sparse Autoencoder Features","created":"2025-08-18","cited_by_count":12,"area":"mechinterp"},{"id":"2505.19398","title":"Erasing Concepts, Steering Generations: A Comprehensive Survey of Concept Suppression","created":"2025-05-26","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":"2510.05081","title":"SAEdit: Token-level control for continuous image editing via Sparse AutoEncoder","created":"2025-10-06","cited_by_count":9,"area":"diffusion"},{"id":"2507.12464","title":"CytoSAE: Interpretable Cell Embeddings for Hematology","created":"2025-07-16","cited_by_count":8,"area":"mechinterp"},{"id":"2602.05027","title":"AudioSAE: Towards Understanding of Audio-Processing Models with Sparse AutoEncoders","created":"2026-02-04","cited_by_count":8,"area":"mechinterp"},{"id":"2603.14819","title":"RAZOR: Ratio-Aware Layer Editing for Targeted Unlearning in Vision Transformers and Diffusion Models","created":"2026-03-16","cited_by_count":8,"area":"mllm"},{"id":"2512.08486","title":"Temporal Concept Dynamics in Diffusion Models via Prompt-Conditioned Interventions","created":"2025-12-09","cited_by_count":7,"area":"diffusion"},{"id":"2603.11493","title":"OrthoEraser: Coupled-Neuron Orthogonal Projection for Concept Erasure","created":"2026-03-12","cited_by_count":7,"area":"mechinterp"},{"id":"2605.00842","title":"Understanding Emergent Misalignment via Feature Superposition Geometry","created":"2026-04-07","cited_by_count":7,"area":"mechinterp"},{"id":"2607.07907","title":"Multimodal Unlearning Across Vision, Language, Video, and Audio: Survey of Methods, Datasets, and Benchmarks","created":"2026-07-08","cited_by_count":7,"area":"safety"},{"id":"2505.24360","title":"Interpreting Large Text-to-Image Diffusion Models with Dictionary Learning","created":"2025-05-30","cited_by_count":6,"area":"diffusion"},{"id":"2504.01137","title":"Follow the Flow: On Information Flow Across Textual Tokens in Text-to-Image Models","created":"2025-04-01","cited_by_count":4,"area":null},{"id":"2507.19894","title":"Generative Model Unlearning: A Survey through Target Events, Unlearning Operators, and Evaluation Protocols","created":"2025-07-26","cited_by_count":4,"area":"safety"},{"id":"2601.00267","title":"ActErase: A Training-Free Paradigm for Precise Concept Erasure via Activation Redirection","created":"2026-01-01","cited_by_count":4,"area":"diffusion"},{"id":"2602.11910","title":"TADA! Tuning Audio Diffusion Models through Activation Steering","created":"2026-02-12","cited_by_count":4,"area":"diffusion"},{"id":"2504.11850","title":"ACE: Attentional Concept Erasure in Diffusion Models","created":"2025-04-16","cited_by_count":3,"area":"diffusion"},{"id":"2602.03410","title":"UnHype: CLIP-Guided Hypernetworks for Dynamic LoRA Unlearning","created":"2026-02-03","cited_by_count":3,"area":"diffusion"},{"id":"2603.03163","title":"Conditioned Activation Transport for T2I Safety Steering","created":"2026-03-03","cited_by_count":3,"area":"mechinterp"},{"id":"2603.15033","title":"Models Designed to Forget: Machine Unlearning via Key Deletion","created":"2026-03-16","cited_by_count":3,"area":"safety"},{"id":"2604.10032","title":"Closed-Form Concept Erasure via Double Projections","created":"2026-04-11","cited_by_count":3,"area":"diffusion"},{"id":"2606.01479","title":"Sparse Autoencoders for Interpretable Emotion Control in Text-to-Speech","created":"2026-05-31","cited_by_count":3,"area":"mechinterp"},{"id":"2606.15796","title":"DifFRACT: Diffusion Feature Reconstruction and Attribution for Circuit Tracing","created":"2026-06-14","cited_by_count":3,"area":"diffusion"},{"id":"2510.03263","title":"Memory Self-Regeneration: Uncovering Hidden Knowledge in Unlearned Models","created":"2025-09-26","cited_by_count":2,"area":"safety"},{"id":"2602.07058","title":"SPARE: Self-distillation for PARameter-Efficient Removal","created":"2026-02-04","cited_by_count":2,"area":"diffusion"},{"id":"2602.07919","title":"Selective Fine-Tuning for Targeted and Robust Concept Unlearning","created":"2026-02-08","cited_by_count":2,"area":"diffusion"},{"id":"2602.10216","title":"ELROND: Exploring and decomposing intrinsic capabilities of diffusion models","created":"2026-02-10","cited_by_count":2,"area":"diffusion"},{"id":"2604.14925","title":"Improving Sparse Autoencoder with Dynamic Attention","created":"2026-04-16","cited_by_count":2,"area":"mechinterp"},{"id":"2605.12122","title":"Disentangled Sparse Representations for Concept-Separated Diffusion Unlearning","created":"2026-05-12","cited_by_count":2,"area":"diffusion"},{"id":"2605.19739","title":"FlowErase-RL: Rethinking Concept Erasure as Reward Optimization in Flow Matching Models","created":"2026-05-19","cited_by_count":2,"area":"diffusion"},{"id":"2609.23658","title":"Why Do Video Diffusion Models Violate Physics? Unveiling the Flaws in Attention Mechanisms","created":"2026-09-20","cited_by_count":2,"area":"diffusion"},{"id":"2509.10809","title":"Rethinking Sparse Autoencoders: Select-and-Project for Fairness and Control from Encoder Features Alone","created":"2025-09-13","cited_by_count":1,"area":"lr"},{"id":"2510.25512","title":"FaCT: Faithful Concept Traces for Explaining Neural Network Decisions","created":"2025-10-29","cited_by_count":1,"area":null},{"id":"2511.01932","title":"Deciphering Personalization: Towards Fine-Grained Explainability in Natural Language for Personalized Image Generation Models","created":"2025-11-02","cited_by_count":1,"area":null},{"id":"2512.15938","title":"SALVE: Sparse Autoencoder-Latent Vector Editing for Mechanistic Control of Neural Networks","created":"2025-12-17","cited_by_count":1,"area":"mechinterp"},{"id":"2601.03213","title":"Critic-Guided Reinforcement Unlearning in Text-to-Image Diffusion","created":"2026-01-06","cited_by_count":1,"area":"diffusion"},{"id":"2605.06610","title":"SoftSAE: Dynamic Top-K Selection for Adaptive Sparse Autoencoders","created":"2026-05-07","cited_by_count":1,"area":"mechinterp"},{"id":"2605.10198","title":"Empty SPACE: Cross-Attention Sparsity for Concept Erasure in Diffusion Models","created":"2026-05-11","cited_by_count":1,"area":"diffusion"},{"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":"2605.27813","title":"Residualized Temporal Sparse Autoencoders for Interpreting Diffusion Models","created":"2026-05-27","cited_by_count":1,"area":"diffusion"},{"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":"2606.31699","title":"Look But Don't Touch with Sparse Autoencoders for Unlearning in Diffusion Models","created":"2026-06-30","cited_by_count":1,"area":"diffusion"},{"id":"2607.08605","title":"When Structured Sparse Autoencoders Learn Consistent Concepts Across Modalities","created":"2026-07-09","cited_by_count":1,"area":"mechinterp"},{"id":"2607.17770","title":"Measuring Monosemanticity in Sparse Autoencoders via Latent Activation Coherence","created":"2026-07-20","cited_by_count":1,"area":"mechinterp"},{"id":"2608.07620","title":"FlowErase-OPD: Multi-Concept Erasure via Anchored On-Policy Distillation in Flow Matching Models","created":"2026-08-07","cited_by_count":1,"area":"diffusion"},{"id":"2608.12829","title":"Semantic Steering for Controllable Generation: Tuning-Free Concept Erasure in Multimodal Diffusion Transformers","created":"2026-08-13","cited_by_count":1,"area":"diffusion"},{"id":"2610.01969","title":"RASteer: Retain-Aware Activation Steering for Concept Erasure in Diffusion Models","created":"2026-10-01","cited_by_count":1,"area":"diffusion"},{"id":"2610.05601","title":"Your Unlearning Gives You Away: Identifying Erased Concepts in Diffusion Models","created":"2026-10-04","cited_by_count":1,"area":"diffusion"},{"id":"2512.10720","title":"Beyond the Black Box: Identifiable Interpretation and Control in Generative Models via Causal Minimality","created":"2025-12-11","cited_by_count":0,"area":null},{"id":"2602.06806","title":"RAIGen: Rare Attribute Identification in Text-to-Image Generative Models","created":"2026-02-06","cited_by_count":0,"area":"diffusion"},{"id":"2603.00975","title":"Forgetting is Competition: Rethinking Unlearning as Representation Interference in Diffusion Models","created":"2026-03-01","cited_by_count":0,"area":"diffusion"},{"id":"2603.07066","title":"PolypSteer: Counterfactual Endoscopic Synthesis via Training-Free Activation Steering","created":"2026-03-07","cited_by_count":0,"area":"diffusion"},{"id":"2604.06063","title":"EDGE-Shield: Efficient Denoising-staGE Shield for Violative Content Filtering via Scalable Reference-Based Matching","created":"2026-04-04","cited_by_count":0,"area":null},{"id":"2605.08218","title":"Deep Dreams Are Made of This: Visualizing Monosemantic Features in Diffusion Models","created":"2026-05-06","cited_by_count":0,"area":"diffusion"},{"id":"2605.15737","title":"BARRIER: Bounded Activation Regions for Robust Information Erasure","created":"2026-05-15","cited_by_count":0,"area":"safety"},{"id":"2605.18719","title":"SafeDiffusion-R1: Online Reward Steering for Safe Diffusion Post-Training","created":"2026-05-18","cited_by_count":0,"area":"diffusion"},{"id":"2605.25574","title":"Mosaic: Compositional Multi-Concept Erasure via Vector Field Blending","created":"2026-05-25","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":"2606.25548","title":"Concept Removal for Frontier Image Generative Models","created":"2026-06-24","cited_by_count":0,"area":null},{"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":"2607.06120","title":"AEGIS: A Mechanism-Guided Defense against Visual Synonym Jailbreaks in Text-to-Image Models","created":"2026-07-07","cited_by_count":0,"area":"diffusion"},{"id":"2607.18516","title":"Signed Rectified Flow: Negativity-Controlled Generation","created":"2026-07-20","cited_by_count":0,"area":"diffusion"},{"id":"2608.14730","title":"IP Protection in the Era of Visual Generative AI: A Survey","created":"2026-08-13","cited_by_count":0,"area":null},{"id":"2609.16229","title":"Test-Time Unlearning via Sparse Autoencoder","created":"2026-09-14","cited_by_count":0,"area":"safety"},{"id":"2609.34137","title":"You Can't Have It Both Ways: Concept Entanglement Limits Diffusion Model Unlearning","created":"2026-09-28","cited_by_count":0,"area":"diffusion"},{"id":"2609.36612","title":"Do LLMs Really Forget? Hidden-State Leakage in Model Unlearning and How to Fix it","created":"2026-09-29","cited_by_count":0,"area":"safety"},{"id":"2609.36832","title":"Motion Concept Unlearning in Video Diffusion Models","created":"2026-09-29","cited_by_count":0,"area":"diffusion"},{"id":"2609.39625","title":"D-Scope: Decomposing and Steering Diffusion Transformers with Sparse Autoencoders","created":"2026-09-30","cited_by_count":0,"area":"diffusion"},{"id":"2610.04020","title":"ClasSAE: Class-Aligned Sparse Autoencoders via Differentiable Feature-Class Affinity","created":"2026-10-02","cited_by_count":0,"area":"mechinterp"},{"id":"2610.06945","title":"Beyond the Linear Representation Hypothesis: Non-Linear Activation Steering in Text-to-Image Models","created":"2026-10-03","cited_by_count":0,"area":"mechinterp"},{"id":"2610.07337","title":"Compositional Concept Erasure in Text-to-Image Diffusion Models via Hierarchically Grounded Semantic Surgery","created":"2026-10-05","cited_by_count":0,"area":"diffusion"},{"id":"2610.13997","title":"SharedSAE: a unified sparse dictionary across u-net blocks and denoising time for diffusion models","created":"2026-09-27","cited_by_count":0,"area":"diffusion"}],"next_cursor":null}