English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

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

Forum topic · 小凯 · 2026-06-27

Summary

This paper introduces a self-evolving training framework that enables unified large multimodal models (LMMs) to improve both visual understanding and image generation autonomously, without human annotations, preference labels, or external reward models. The framework assigns three internal roles within a single model: a Proposer that generates visual questions from unlabeled images, a Solver that answers and evaluates them, and a Generator that produces images, with self-consistency rewards serving as the training signal. By leveraging only unlabeled images, the approach removes the dependency on curated post-training supervision. Reported on arXiv as 2606.27376, the work targets the computer vision research community interested in unified multimodal modeling and self-supervised improvement loops.

Paper Overview

Research Area: Computer Vision (CV)

Authors: Ritesh Thawkar, Shravan Venkatraman, Omkar Thawakar

Published: 2026-06-27

arXiv: 2606.27376

Abstract

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. This paper asks whether a unified LMM can improve both abilities autonomously using only unlabeled images.

The authors propose a self-evolving training framework built around three internal roles:

  • Proposer — generates visual questions from unlabeled images
  • Solver — answers and evaluates those questions
  • Generator — produces images, guided by self-consistency rewards
  • By using self-consistency as the reward signal, the framework allows the model to supervise its own improvement across both understanding and generation tasks, removing the need for external reward models or human-labeled preference data.

    Links

  • arXiv page: https://arxiv.org/abs/2606.27376
*Auto-collected on 2026-06-27.*

Tags

#paper#arxiv#computer-vision#multimodal#large-multimodal-models#self-supervised-learning#image-generation#self-consistency

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178208189