{
  "annotation_id": "324d69ed-e05a-4792-8753-11364cee5134",
  "date_created": "2026-02-17T05:53:12.655000Z",
  "date_modified": "2026-02-17T05:53:12.655000Z",
  "file_hash": "649e3b9edc275b8f04b323718c75c7f439bf84eb6be70d295c48aa729015bcfb",
  "private": false,
  "record": {
    "abstract": "While inference-time scaling has significantly enhanced generative quality in large language and diffusion models, its application to vector-quantized (VQ) visual autoregressive modeling (VAR) remains unexplored. We introduce VAR-Scaling, the first general framework for inference-time scaling in VAR, addressing the critical challenge of discrete latent spaces that prohibit continuous path search. We find that VAR scales exhibit two distinct pattern types: general patterns and specific patterns, where later-stage specific patterns conditionally optimize early-stage general patterns. To overcome the discrete latent space barrier in VQ models, we map sampling spaces to quasi-continuous feature spaces via kernel density estimation (KDE), where high-density samples approximate stable, high-quality solutions. This transformation enables effective navigation of sampling distributions. We propose a density-adaptive hybrid sampling strategy: Top-k sampling focuses on high-density regions to preserve quality near distribution modes, while Random-k sampling explores low-density areas to maintain diversity and prevent premature convergence. Consequently, VAR-Scaling optimizes sample fidelity at critical scales to enhance output quality. Experiments in class-conditional and text-to-image evaluations demonstrate significant improvements in inference process. The code is available at https://github.com/WD7ang/VAR-Scaling.",
    "arxiv_id": "2601.07293",
    "authors": [
      "Weidong Tang",
      "Xinyan Wan",
      "Siyu Li",
      "Xiumei Wang"
    ],
    "categories": [
      "cs.CV"
    ],
    "doi": "10.1007/978-981-95-5699-1_28",
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "title": "Inference-Time Scaling for Visual AutoRegressive modeling by Searching Representative Samples",
    "url": "https://arxiv.org/abs/2601.07293",
    "version": "v1"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "79c1dac3-ca2d-4b76-9db4-0e260e929c1e",
    "id": "arXiv Dataset",
    "type": "Model",
    "variant": "snapshot-2026-01-17",
    "version": "0.1.0"
  },
  "user_id": 1000002
}