生成式通用验证器作为多模态元推理器

Generative Universal Verifier as Multimodal Meta-Reasoner

October 15, 2025

作者: Xinchen Zhang, Xiaoying Zhang, Youbin Wu, Yanbin Cao, Renrui Zhang, Ruihang Chu, Ling Yang, Yujiu Yang

cs.AI

摘要

我们提出了生成式通用验证器这一创新概念与插件，专为下一代视觉-语言模型及统一多模态模型中的多模态推理设计，提供了在推理与生成过程中对视觉结果进行反思与优化的基础能力。本研究主要贡献有三：(1) 构建了ViVerBench，一个涵盖16类关键任务的综合基准，用于评估多模态推理中的视觉输出。结果显示，现有视觉-语言模型在这些任务上普遍表现不佳，凸显了与人类可靠视觉验证能力间的显著差距。(2) 设计了两条自动化流水线，用于构建大规模视觉验证数据并训练OmniVerifier-7B，这是首个具备全方位能力的生成式验证器，专为通用视觉验证而训练，在ViVerBench上取得了显著提升(+8.3)。通过训练，我们识别出视觉验证中的三项基本能力，并展示了它们如何协同泛化与交互。(3) 提出了OmniVerifier-TTS，一种序列化测试时扩展范式，利用通用验证器在统一模型内桥接图像生成与编辑，通过迭代细粒度优化提升生成能力的上限。除生成外，我们将通用验证器扩展至更广泛的世界模型交织推理场景。实证表明，OmniVerifier-TTS在T2I-ReasonBench(+3.7)和GenEval++(+4.3)上取得改进，超越了如Best-of-N等现有并行测试时扩展方法。通过赋予多模态推理可靠的视觉验证能力，OmniVerifier推动了生成过程中的可靠反思与可扩展的测试时优化，标志着向更可信、可控的下一代推理系统迈进了一步。

English

We introduce Generative Universal Verifier, a novel concept and plugin designed for next-generation multimodal reasoning in vision-language models and unified multimodal models, providing the fundamental capability of reflection and refinement on visual outcomes during the reasoning and generation process. This work makes three main contributions: (1) We build ViVerBench, a comprehensive benchmark spanning 16 categories of critical tasks for evaluating visual outcomes in multimodal reasoning. Results show that existing VLMs consistently underperform across these tasks, underscoring a substantial gap from human-level capability in reliable visual verification. (2) We design two automated pipelines to construct large-scale visual verification data and train OmniVerifier-7B, the first omni-capable generative verifier trained for universal visual verification and achieves notable gains on ViVerBench(+8.3). Through training, we identify three atomic capabilities in visual verification and demonstrate how they generalize and interact synergistically. (3) We propose OmniVerifier-TTS, a sequential test-time scaling paradigm that leverages the universal verifier to bridge image generation and editing within unified models, enhancing the upper bound of generative ability through iterative fine-grained optimization. Beyond generation, we extend universal verifier to broader world-modeling interleaved reasoning scenarios. Empirically, OmniVerifier-TTS achieves improvements on T2I-ReasonBench(+3.7), and GenEval++(+4.3), outperforming existing parallel test-time scaling methods, such as Best-of-N. By endowing multimodal reasoning with reliable visual verification, OmniVerifier advances both reliable reflection during generation and scalable test-time refinement, marking a step toward more trustworthy and controllable next-generation reasoning systems.