UniGen-1.5: Enhancing Image Generation and Editing through Reward Unification in Reinforcement Learning
This work addresses the need for improved image generation and editing capabilities in AI, though it appears incremental as it builds upon the previous UniGen model.
The paper tackles the problem of enhancing image generation and editing by proposing UniGen-1.5, a multimodal large language model that uses a unified reinforcement learning strategy with shared reward models, achieving scores of 0.89 on GenEval and 4.31 on ImgEdit, surpassing state-of-the-art models like BAGEL.
We present UniGen-1.5, a unified multimodal large language model (MLLM) for advanced image understanding, generation and editing. Building upon UniGen, we comprehensively enhance the model architecture and training pipeline to strengthen the image understanding and generation capabilities while unlocking strong image editing ability. Especially, we propose a unified Reinforcement Learning (RL) strategy that improves both image generation and image editing jointly via shared reward models. To further enhance image editing performance, we propose a light Edit Instruction Alignment stage that significantly improves the editing instruction comprehension that is essential for the success of the RL training. Experimental results show that UniGen-1.5 demonstrates competitive understanding and generation performance. Specifically, UniGen-1.5 achieves 0.89 and 4.31 overall scores on GenEval and ImgEdit that surpass the state-of-the-art models such as BAGEL and reaching performance comparable to proprietary models such as GPT-Image-1.