HarMoE: Multi-Source Chest Radiograph Pretraining with Dataset-Disentangled Experts
Recent vision-language models for chest X-ray understanding are largely built on image-report alignment and therefore rely heavily on MIMIC-CXR as the dominant pretraining source. While effective at scale, this paradigm underexplores an important alternative source of supervision: a range of existing multi-label classification datasets, which provide cleaner and more explicit disease signals than free-text reports, and can offer broader pathology coverage when combined across sources. However, l
Record details
Published: 3 August 2026
Source: arXiv
Category: Research
Topics: Safety & alignment
Retrieved: 4 August 2026
Related evidence
These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.
From Profiling to Synthesis: Benchmarking Implicit Behavioral Alignment in Personalized LLM Agents
arXiv · 3 August 2026
An AI-Based Decision-Support Pipeline for Day-Ahead Photovoltaic Forecasting
arXiv · 3 August 2026
Advancing Relevance Measurement with Vision-Language Models for Web-Scale Search
arXiv cs.LG · 3 August 2026
STEAM:ASpatio-TEmporal Alignment Mixture-of-Experts Model with Hierarchical Pre-training for EEG Decoding
arXiv cs.LG · 3 August 2026
Stuck on "A": Diagnosing and Repairing Interface Injury in Attention-to-KDA Linearization of a 0.6B Language Model
arXiv cs.LG · 3 August 2026
Learning Molecular Representations from Cellular Phenotypes with Structure Preservation
arXiv cs.LG · 3 August 2026
How to cite this record
ethics.ai (3 August 2026), “HarMoE: Multi-Source Chest Radiograph Pretraining with Dataset-Disentangled Experts,” evidence record 15879, https://ethics.ai/record/15879 (originally published by arXiv).
Use and limitations
This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.