[
    {
        "id": "osp-17756",
        "type": "article-journal",
        "title": "QuadTok: Quadtree Visual Tokenizer for Autoregressive Image Generation",
        "author": [
            {
                "family": "Mao",
                "given": "Yucheng"
            },
            {
                "family": "Chen",
                "given": "Zeyuan"
            },
            {
                "family": "Shan",
                "given": "Xiaojun"
            },
            {
                "family": "Zhang",
                "given": "Xiang"
            },
            {
                "family": "Srivastava",
                "given": "Divyansh"
            },
            {
                "family": "Li",
                "given": "Bingnan"
            },
            {
                "family": "Tu",
                "given": "Zhuowen"
            }
        ],
        "URL": "https://omanscience.com/en/articles/quadtok-quadtree-visual-tokenizer-for-autoregressive-image-generation",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "We introduce QuadTok, a novel framework for visual tokenization and autoregressive image generation. Compared to traditional approaches using 2D grids or 1D token sequences, we propose a hierarchical quadtree structure, bridging the gap between 2D spatial binding and 1D sequence-level flexibility. The QuadTok tokenizer dynamically allocates representational capacity to visually intricate areas while leaving homogeneous regions at a coarse resolution. Compared with a fixed 256-token grid, our ImageNet-trained tokenizer saves approximately 10% of tokens on ImageNet and 9% when transferred zero-shot to the COCO dataset, while maintaining comparable reconstruction fidelity. Furthermore, the natural causality introduced by the tree structure seamlessly enables autoregressive image generation. Conditioned on a quadtree topology supplied before generation, our 947M GPT-style generative model achieves a 2.08 gFID on the ImageNet $256 \\times 256$ benchmark. Additionally, leveraging the strong spatial correlation preserved by the quadtree structure, the QuadTok generator enables zero-shot spatially controlled image generation capabilities. Code: https://github.com/myc634/QuadTok."
    }
]