Calibrated Confidence Expression for Radiology Report Generation
ConRad (Confidence Calibration for Radiology Reports), a reinforcement learning framework for fine-tuning medical LVLMs to produce calibrated verbalized confidence estimates alongside radiology reports, is introduced and substantially improves calibration and outperforms competing methods.
D. Bani-Harouni, Chantal Pellegrini, Julian Lüers et al.
· arXiv.org · 0 citations