Research

Paper

AI LLM March 04, 2026

BLOCK: An Open-Source Bi-Stage MLLM Character-to-Skin Pipeline for Minecraft

Authors

Hengquan Guo

Abstract

We present \textbf{BLOCK}, an open-source bi-stage character-to-skin pipeline that generates pixel-perfect Minecraft skins from arbitrary character concepts. BLOCK decomposes the problem into (i) a \textbf{3D preview synthesis stage} driven by a large multimodal model (MLLM) with a carefully designed prompt-and-reference template, producing a consistent dual-panel (front/back) oblique-view Minecraft-style preview; and (ii) a \textbf{skin decoding stage} based on a fine-tuned FLUX.2 model that translates the preview into a skin atlas image. We further propose \textbf{EvolveLoRA}, a progressive LoRA curriculum (text-to-image $\rightarrow$ image-to-image $\rightarrow$ preview-to-skin) that initializes each phase from the previous adapter to improve stability and efficiency. BLOCK is released with all prompt templates and fine-tuned weights to support reproducible character-to-skin generation.

Metadata

arXiv ID: 2603.03964
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-04
Fetched: 2026-03-05 06:06

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.03964v1</id>\n    <title>BLOCK: An Open-Source Bi-Stage MLLM Character-to-Skin Pipeline for Minecraft</title>\n    <updated>2026-03-04T11:55:32Z</updated>\n    <link href='https://arxiv.org/abs/2603.03964v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.03964v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>We present \\textbf{BLOCK}, an open-source bi-stage character-to-skin pipeline that generates pixel-perfect Minecraft skins from arbitrary character concepts. BLOCK decomposes the problem into (i) a \\textbf{3D preview synthesis stage} driven by a large multimodal model (MLLM) with a carefully designed prompt-and-reference template, producing a consistent dual-panel (front/back) oblique-view Minecraft-style preview; and (ii) a \\textbf{skin decoding stage} based on a fine-tuned FLUX.2 model that translates the preview into a skin atlas image. We further propose \\textbf{EvolveLoRA}, a progressive LoRA curriculum (text-to-image $\\rightarrow$ image-to-image $\\rightarrow$ preview-to-skin) that initializes each phase from the previous adapter to improve stability and efficiency. BLOCK is released with all prompt templates and fine-tuned weights to support reproducible character-to-skin generation.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-04T11:55:32Z</published>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Hengquan Guo</name>\n    </author>\n  </entry>"
}