蚂蚁集团提出 ARGenSeg-8B：基于自回归图像生成模型的图像分割框架

inclusionAI/ARGenSeg-8B

AI 导读

蚂蚁集团推出 ARGenSeg-8B，一种将多模态理解与像素级感知统一的自回归图像生成分割框架。它利用多模态大语言模型（MLLM）输出视觉 token，并通过通用 VQ-VAE 解码为分割掩码，使分割完全依赖 MLLM 的像素级理解。采用 next-scale-prediction 策略并行生成视觉 token，降低推理延迟。在多个分割数据集上超越此前最优方法，推理速度显著提升。论文已被 NeurIPS 2025 接收，模型已发布在 HuggingFace。

推荐理由

蚂蚁提出用自回归生成做分割，把理解和像素级感知统一到一个框架里，多个数据集SOTA且速度更快，做CV的值得看看。

正文

ARGenSeg: Image Segmentation with Autoregressive Image Generation Model

1 Ant Group · NeurIPS 2025

🏠 About

We propose a novel AutoRegressive Generation-based paradigm for image Segmentation (ARGenSeg), achieving multimodal understanding and pixel-level perception within a unified framework. Prior works integrating image segmentation into multimodal large language models (MLLMs) typically employ either boundary points representation or dedicated segmentation heads. These methods rely on discrete representations or semantic prompts fed into task-specific decoders, which limits the ability of the MLLM to capture fine-grained visual details. To address these challenges, we introduce a segmentation framework for MLLM based on image generation, which naturally produces dense masks for target objects. We leverage MLLM to output visual tokens and detokenize them into images using an universal VQ-VAE, making the segmentation fully dependent on the pixel-level understanding of the MLLM. To reduce inference latency, we employ a next-scale-prediction strategy to generate required visual tokens in parallel. Extensive experiments demonstrate that our method surpasses prior state-of-the-art approaches on multiple segmentation datasets with a remarkable boost in inference speed, while maintaining strong understanding capabilities.

🔗 Citation

If you find this work useful, please cite:

@article{wang2025argenseg,
  title={ARGenSeg: Image Segmentation with Autoregressive Image Generation Model},
  author={Wang, Xiaolong and Ru, Lixiang and Huang, Ziyuan and Ji, Kaixiang and Zheng, Dandan and Chen, Jingdong and Zhou, Jun},
  journal={arXiv preprint arXiv:2510.20803},
  year={2025}
}

👏 Acknowledgements

We sincerely thank the contributors of InternVL, VAR, and PSALM for their foundational work and open-source spirit.