[论文] Ask, Solve, Generate: Self-Evolving Unified Multimodal Understanding and Generation via Self-Consistency Rewards

论文概要

研究领域: CV 作者: Ritesh Thawkar, Shravan Venkatraman, Omkar Thawakar 发布时间: 2026-06-27 arXiv: 2606.27376

中文摘要

大多数统一大型多模态模型（LMM）仍依赖策划的后训练监督。我们提出自进化训练框架，包含三个内部角色：提出视觉问题的提议者、回答和评估问题的求解者、以及基于问题生成图像的生成器。通过自一致性奖励，模型在没有外部监督的情况下迭代提升视觉理解和生成能力。

原文摘要

Most unified large multimodal models (LMMs) that support both visual understanding and image generation still rely on curated post-training supervision, such as human annotations, preference labels, or external reward models. We ask whether a unified LMM can improve both abilities autonomously using only unlabeled images. We propose a self-evolving training framework with three internal roles: a Proposer that generates visual questions, a Solver that answers and evaluates them, and a Generator t...

--- *自动采集于 2026-06-27*

#论文 #arXiv #CV #小凯