paper-with-me

홈 › Papers

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

2025-04-03 · Xianwei Zhuang, Yuxin Xie, Yufan Deng, Dongchao Yang, Liming Liang, Jinghan Ru, Yuguo Yin, Yuexian Zou

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.

📄 PDF Abstract BibTeX arXiv:2504.02949

Code (2)

vargpt-family/vargpt-v1.1 공식 구현 pytorch
VARGPT-family/VARGPT pytorch

Tasks

Image GenerationInstruction Following

Methods 이 논문이 사용한 방법론

ADOPT Please enter a description about the method here

Similar Papers 제목 키워드 기반

VARGPT: Unified Understanding and Generation in a Visual Autoregressive Multimodal Large Language Model

2025-01-21 · Xianwei Zhuang, Yuxin Xie, Yufan Deng, Liming Liang 외

We present VARGPT, a novel multimodal large language model (MLLM) that unifies visual understanding and generation within a single autoregressive framework. VARGPT employs a next-token prediction paradigm for visual unde…

Image GenerationInstruction FollowingLanguage ModelingLanguage Modelling+4

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

2026-06-25 · Ritesh Thawkar, Shravan Venkatraman, Omkar Thawakar, Abdelrahman Shaker 외 arxiv

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 mo…

Image Generation

Unified Autoregressive Visual Generation and Understanding with Continuous Tokens

2025-03-17 · Lijie Fan, Luming Tang, Siyang Qin, Tianhong Li 외

We present UniFluid, a unified autoregressive framework for joint visual generation and understanding leveraging continuous visual tokens. Our unified autoregressive architecture processes multimodal image and text input…

Image CaptioningImage GenerationQuestion Answering

Unified Multimodal Autoregressive Modeling with Shared Context-Visual Tokenizer is Key to Unification

2026-06-16 · Wujian Peng, Lingchen Meng, Yuxuan Cai, Xianwei Zhuang 외 arxiv

Unified Multimodal Modeling aims to integrate visual understanding and generation within a single system. However, existing approaches typically rely on two disparate visual tokenizers, which splits the representation sp…

Reinforcement LearningImage GenerationImage Editing

Growing Visual Generative Capacity for Pre-Trained MLLMs

2025-10-02 · Hanyu Wang, Jiaming Han, Ziyan Yang, Qi Zhao 외 arxiv

Multimodal large language models (MLLMs) extend the success of language models to visual understanding, and recent efforts have sought to build unified MLLMs that support both understanding and generation. However, const…