VARGPT-v1.1: Improve Visual Autoregressive Large Unified Model via Iterative Instruction Tuning and Reinforcement Learning
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.
Code (2)
Tasks
Image GenerationInstruction FollowingMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
VARGPT: Unified Understanding and Generation in a Visual Autoregressive Multimodal Large Language Model
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+4Ask, Solve, Generate: Self-Evolving Unified Multimodal Understanding and Generation via Self-Consistency Rewards
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 GenerationUnified Autoregressive Visual Generation and Understanding with Continuous Tokens
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 AnsweringUnified Multimodal Autoregressive Modeling with Shared Context-Visual Tokenizer is Key to Unification
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 EditingGrowing Visual Generative Capacity for Pre-Trained MLLMs
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…