sparse autoencoder

Interpretable Safety Alignment via SAE-Constructed Low-Rank Subspace Adaptation ACL 2026 LangSAE Editing: Improving Multilingual Information Retrieval via Post-hoc Language Identity Removal ACL 2026 Language Models Can Explain Visual Features via Steering CVPR 2026 Leveraging Sparsity for Privacy in Collaborative Inference WACV 2026 Measuring Mechanistic Independence: Can Bias Be Removed Without Erasing Demographics? EACL 2026 Mechanistic Interpretability Should Prioritize Feature Consistency in Sparse Autoencoders ACL 2026 Multilingual Language Models Encode Script Over Linguistic Structure ACL 2026 Out of Distribution, Out of Luck: Process Rewards Misguide Reasoning Models EACL 2026 Probing Bias Formation in Medical LLMs through Activation Steering ACL 2026 ProtSAE: Disentangling and Interpreting Protein Language Models via Semantically-Guided Sparse Autoencoders AAAI 2026 SAGE: An Agentic Explainer Framework for Interpreting SAE Features in Language Models EACL 2026 SAVE: Sparse Autoencoder-Driven Visual Information Enhancement for Mitigating Object Hallucination WACV 2026 SCOPE: Intrinsic Semantic Space Control for Mitigating Copyright Infringement in LLMs AAAI 2026 SegCompass: Exploring Interpretable Alignment with Sparse Autoencoders for Enhanced Reasoning Segmentation CVPR 2026 Semantically Comprehensive Token Pruning in LVLMs via Maximizing Concept Coverage ACL 2026 Sparse Feature Coactivation Reveals Causal Semantic Modules in Large Language Models ACL 2026 SparseRM: A Lightweight Preference Modeling with Sparse Autoencoder AAAI 2026 Sparsity as a Key: Unlocking New Insights from Latent Structures for Out-of-Distribution Detection CVPR 2026 Steering Large Language Models for Machine Translation Personalization EACL 2026 Steering Sparse Autoencoder Latents to Control Dynamic Head Pruning in Vision Transformers (Student Abstract) AAAI 2026 Stop Probing, Start Coding: Why Linear Probes and Sparse Autoencoders Fail at Compositional Generalization UAI 2026 Textual Steering Vectors Can Improve Visual Understanding in Multimodal Large Language Models ACL 2026 The Unintended Trade-off of AI Alignment: Balancing Hallucination Mitigation and Safety in LLMs EACL 2026 TIDE: Temporal-Aware Sparse Autoencoders for Interpretable Diffusion Transformers in Image Generation AAAI 2026 Uncovering Sentiment Analysis Circuit in Large Language Model ACL 2026