VARGPT-v1.1: Improve Visual Autoregressive Large Unified Model via
  Iterative Instruction Tuning and Reinforcement Learning

Xianwei Zhuang; Yuxin Xie; Yufan Deng; Dongchao Yang; Liming Liang; Jinghan Ru; Yuguo Yin; Yuexian Zou

返回文献探索

Paper · arXiv 2504.02949

VARGPT-v1.1: Improve Visual Autoregressive Large Unified Model via Iterative Instruction Tuning and Reinforcement Learning

Xianwei Zhuang, Yuxin Xie, Yufan Deng, Dongchao Yang, Liming Liang, Jinghan Ru, Yuguo Yin, Yuexian Zou

21 upvotesApril 3, 2025arXiv 预印本

AI 摘要

VARGPT-v1.1, an advanced visual autoregressive model, integrates iterative visual instruction tuning with reinforcement learning, enhances resolution, and achieves state-of-the-art performance in multimodal understanding and text-to-image tasks.

visual autoregressive modelnext-token predictionnext-scale generationiterative visual instruction tuningreinforcement learningDirect Preference Optimizationexpanded training corpusQwen2enhanced image generation resolutionemergent image editing capabilitiesmultimodal understandingtext-to-image instruction-following taskslarge language models

Abstract

In this work, we present VARGPT-v1.1, an advanced unified visual autoregressive model that builds upon our previous framework VARGPT. The model preserves the dual paradigm of next-token prediction for visual understanding and next-scale generation for image synthesis. Specifically, VARGPT-v1.1 integrates: (1) a novel training strategy combining iterative visual instruction tuning with reinforcement learning through Direct Preference Optimization (DPO), (2) an expanded training corpus containing 8.3M visual-generative instruction pairs, (3) an upgraded language model backbone using Qwen2, (4) enhanced image generation resolution, and (5) emergent image editing capabilities without architectural modifications. These advancements enable VARGPT-v1.1 to achieve state-of-the-art performance in multimodal understanding and text-to-image instruction-following tasks, demonstrating significant improvements in both comprehension and generation metrics. Notably, through visual instruction tuning, the model acquires image editing functionality while maintaining architectural consistency with its predecessor, revealing the potential for unified visual understanding, generation, and editing. Our findings suggest that well-designed unified visual autoregressive models can effectively adopt flexible training strategies from large language models (LLMs), exhibiting promising scalability. The codebase and model weights are publicly available at https://github.com/VARGPT-family/VARGPT-v1.1.