PaDoc: Layout-Grounded Parallel Decoding for Document Parsing

• 类型：arxiv
• 标识：2608.06146
• 链接：https://arxiv.org/abs/2608.06146
• 主分类：multimodal
• 形态：method
• 被引：0
• 被引来源：Semantic Scholar
• S2被引：0
• 影响力被引：0
• TLDR：PaDoc is proposed, a layout-grounded parser that treats the predicted layout as a branching structure over a shared page representation that is the fastest end-to-end parser at five concurrency levels and is the fastest end-to-end parser at five concurrency levels.
• 待LLM分类：否
• 成熟度：research
• 场景：document parsing、layout understanding、parallel decoding
• 标题中文：PaDoc：面向文档解析的布局引导并行解码
• TLDR中文：PaDoc 是一个布局驱动的解析器，将预测布局视为共享页面表示上的分支结构，在五个并发级别下均为最快的 end-to-end 解析器。
• 来源文件：
• /inbox/tom/_candidates/2026-08-07-agent-rag-longcontext-candidates.json
• [S2 enrich]