Clinical Concept Centers in LLMs
cs.CL, cs.LG
Submitted: 2026-10-02
Updated: 2026-10-05
Code: https://github.com/decoderesearch/SAELens
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- An X-Ray Is Worth 15 Features: Sparse Autoencoders for Interpretable Radiology Report Generation
- Understanding intermediate layers using linear classifier probes
- HealthBench: Evaluating Large Language Models Towards Improved Human Health
- A is for Absorption: Studying Feature Splitting and Absorption in Sparse Autoencoders
- Qwen-Scope: Turning Sparse Features into Development Tools for Large Language Models
- Decomposing The Dark Matter of Sparse Autoencoders
- Where Do Apparent LLM Clinical Triage Failures Arise? Localizing the Multiple-Choice Format Effect
- Are Sparse Autoencoders Useful? A Case Study in Sparse Probing
- Evaluating Sparse Autoencoders on Targeted Concept Erasure Tasks
- The Remarkable Robustness of LLMs: Stages of Inference?
- Measuring Faithfulness in Chain-of-Thought Reasoning
- Mechanistic Interpretability of EEG Foundation Models via Sparse Autoencoders
- A Monosemantic Attribution Framework for Stable Interpretability in Clinical Neuroscience Transformer-Based Language Models
- CLR-voyance: Reinforcing Open-Ended Reasoning for Inpatient Clinical Decision Support with Outcome-Aware Rubrics
- Resurrecting the Salmon: Rethinking Mechanistic Interpretability with Domain-Specific Sparse Autoencoders
- Sparse Autoencoders Trained on the Same Data Learn Different Features
- Automatically Interpreting Millions of Features in Large Language Models
- Jumping Ahead: Improving Reconstruction Fidelity with JumpReLU Sparse Autoencoders
- MedGemma Technical Report
- Explaining black box text modules in natural language with language models
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering