aiqbench analysis
This research addresses data scarcity and bias in medical vision-language models by introducing SemEnrich, a self-supervised method that semantically clusters radiology reports to augment training datasets with omitted positive or neutral findings. The approach demonstrates significant performance improvements across multiple evaluation metrics without relying on random augmentation, while also enhancing reward design for GRPO training. By improving the quality and balance of medical imaging data, this work directly advances the reliability of AI systems in clinical diagnostic settings.
SemEnrich: Self-Supervised Semantic Enrichment of Radiology Reports for Vision-Language Learning
arXiv:2604.09887v2 Announce Type: replace Abstract: Medical vision-language datasets are often limited in size and biased toward negative findings, as clinicians report abnormalities mostly but might omit some positive/neutral findings because they might be considered as irrelevant to the patient's condition. We propose a self-supervised data enrichment method that leverages semantic clustering of report sentences. Then we enrich the findings in the medical reports in the training set by adding positive/neutral observations from different clusters in a self-supervised manner. Our approach yields consistent gains in supervised fine-tuning (5.63%, 3.04%, 7.40%, 5.30%, 7.47% average gains on COMET score, Bert score, Sentence Bleu, CheXbert-F1 and RadGraph-F1 scores respectively). Ablation studies confirm that improvements stem from semantic clustering rather than random augmentation. Furthermore, we introduce a way to incorporate semantic cluster information into the reward design for GRPO training, which leads to further performance gains (2.78%, 3.14%, 12.80% average gains on COMET score, Bert score and Sentence Bleu scores respectively). We share our code at https://github.com/igulluk/SemEnrich
Abstract:Medical vision-language datasets are often limited in size and biased toward negative findings, as clinicians report abnormalities mostly but might omit some positive/neutral findings because they might be considered as irrelevant to the patient's condition. We propose a self-supervised data enrichment method that leverages semantic clustering of report sentences. Then we enrich the findings in the medical reports in the training set by adding positive/neutral observations from different clusters in a self-supervised manner. Our approach yields consistent gains in supervised fine-tuning (5.63%, 3.04%, 7.40%, 5.30%, 7.47% average gains on COMET score, Bert score, Sentence Bleu, CheXbert-F1 and RadGraph-F1 scores respectively). Ablation studies confirm that improvements stem from semantic clustering rather than random augmentation. Furthermore, we introduce a way to incorporate semantic cluster information into the reward design for GRPO training, which leads to further performance gains (2.78%, 3.14%, 12.80% average gains on COMET score, Bert score and Sentence Bleu scores respectively). We share our code at this https URL
Machine Learning (cs.LG)
arXiv:2604.09887 [cs.LG]
(or arXiv:2604.09887v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2604.09887 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 251 words
Related coverage
TRACE: Training-time Report-guided and Clinically Ordered Concept Editing
arXiv · Aug 24, 2026
TextSLIP: Text Self-Supervised CLIP for Medical Report Generation
arXiv · Jul 28, 2026
GLINT: Sparsely Gated Vision-Language Alignment for Fine-Grained Radiology Representations
arXiv · Jun 3, 2026
Biomedical Imaging's Real Bottleneck Is the Data, Not the Model
Databricks · Oct 8, 2026