Search Results for author:
Found 2 papers, 1 papers with code
An X-Ray Is Worth 15 Features: Sparse Autoencoders for Interpretable Radiology Report Generation
Using an off-the-shelf language model, we distil ground-truth reports into radiological descriptions for each SAE feature, which we then compile into a full report for each image, eliminating the need for fine-tuning large models for this task.
Comparing Optimization Targets for Contrast-Consistent Search
We investigate the optimization target of Contrast-Consistent Search (CCS), which aims to recover the internal representations of truth of a large language model.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
