无需训练即可加速文档解析模型，提升推理速度同时保持全局一致性。
HSD: Training-Free Acceleration for Document Parsing Vision-Language Models with Hierarchical Speculative Decoding
- 分层推测解码：先局部粗略生成，再全局验证
- 在OmniDocBench上实现2.7倍加速，长文档达7.04倍
- 适合需要快速解析长文档的工业级应用
文档解析是多模态理解的基础任务，支持信息提取与智能文档分析等广泛应用。近年来，基于视觉语言模型（VLM）的端到端方法成为主流，但其推理延迟高，需自回归生成长页面序列。现有混合方法虽通过区域级并行解码缓解延迟，但独立区域解码会丢失全页上下文，影响整体连贯性。为此，我们提出分层推测解码（HSD），一种两阶段从局部到全局的框架。第一阶段使用轻量级流水线起草器预测区域划分并生成各区域粗略草稿；第一阶段并行验证草稿以提升效率，第二阶段对优化后输出进行页面级验证，以保持全页一致性。实验表明，HSD在HunyuanOCR上实现近无损2.7倍加速，在长文档解析任务中最高达7.04倍加速，验证了方法有效性。代码已开源：https://github.com/whlscut/HSD。
原文摘要 · Abstract (English)
Document parsing is a fundamental task in multimodal understanding, supporting a wide range of downstream applications such as information extraction and intelligent document analysis. Benefiting from strong semantic modeling and robust generalization, VLM-based end-to-end approaches have emerged as the mainstream paradigm in recent years. However, these models often suffer from substantial inference latency, as they must autoregressively generate long, full-page sequences when processing long-form documents. While recent hybrid methods mitigate this issue via region-level parallel decoding with VLMs, independent region decoding loses full-page context and might weaken global coherence. To address this issue, we propose Hierarchical Speculative Decoding (HSD), a two-stage local-to-global framework for document parsing. HSD first employs a lightweight pipeline drafter to predict region partitions and generate coarse drafts for each region. The first stage verifies the generated region-level drafts in parallel for efficiency, while the second stage further performs page-level verification on these refined outputs to preserve full-page coherence. Experimental results show that HSD achieves a near-lossless 2.7x speedup with HunyuanOCR on OmniDocBench v1.5 and up to 7.04x speedup on long-document parsing tasks, demonstrating the effectiveness of the proposed method. The code is available at https://github.com/whlscut/HSD.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。