Random Direct Preference Optimization for Radiography Report Generation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Samokhin, Valentin, Shirokikh, Boris, Goncharov, Mikhail, Umerenkov, Dmitriy, Bobrin, Maksim, Oseledets, Ivan, Dylov, Dmitry, Belyaev, Mikhail
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914057046458368
author Samokhin, Valentin
Shirokikh, Boris
Goncharov, Mikhail
Umerenkov, Dmitriy
Bobrin, Maksim
Oseledets, Ivan
Dylov, Dmitry
Belyaev, Mikhail
author_facet Samokhin, Valentin
Shirokikh, Boris
Goncharov, Mikhail
Umerenkov, Dmitriy
Bobrin, Maksim
Oseledets, Ivan
Dylov, Dmitry
Belyaev, Mikhail
contents Radiography Report Generation (RRG) has gained significant attention in medical image analysis as a promising tool for alleviating the growing workload of radiologists. However, despite numerous advancements, existing methods have yet to achieve the quality required for deployment in real-world clinical settings. Meanwhile, large Visual Language Models (VLMs) have demonstrated remarkable progress in the general domain by adopting training strategies originally designed for Large Language Models (LLMs), such as alignment techniques. In this paper, we introduce a model-agnostic framework to enhance RRG accuracy using Direct Preference Optimization (DPO). Our approach leverages random contrastive sampling to construct training pairs, eliminating the need for reward models or human preference annotations. Experiments on supplementing three state-of-the-art models with our Random DPO show that our method improves clinical performance metrics by up to 5%, without requiring any additional training data.
format Preprint
id arxiv_https___arxiv_org_abs_2509_21351
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Random Direct Preference Optimization for Radiography Report Generation
Samokhin, Valentin
Shirokikh, Boris
Goncharov, Mikhail
Umerenkov, Dmitriy
Bobrin, Maksim
Oseledets, Ivan
Dylov, Dmitry
Belyaev, Mikhail
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Radiography Report Generation (RRG) has gained significant attention in medical image analysis as a promising tool for alleviating the growing workload of radiologists. However, despite numerous advancements, existing methods have yet to achieve the quality required for deployment in real-world clinical settings. Meanwhile, large Visual Language Models (VLMs) have demonstrated remarkable progress in the general domain by adopting training strategies originally designed for Large Language Models (LLMs), such as alignment techniques. In this paper, we introduce a model-agnostic framework to enhance RRG accuracy using Direct Preference Optimization (DPO). Our approach leverages random contrastive sampling to construct training pairs, eliminating the need for reward models or human preference annotations. Experiments on supplementing three state-of-the-art models with our Random DPO show that our method improves clinical performance metrics by up to 5%, without requiring any additional training data.
title Random Direct Preference Optimization for Radiography Report Generation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2509.21351