SynthDocBench: Controlled Benchmark for Long-Context Visual Document Understanding

  • 类型：arxiv
  • 标识：2607.10400
  • 链接：https://arxiv.org/abs/2607.10400
  • 主分类：evaluation
  • 形态：benchmark
  • 被引：0
  • 被引来源：Semantic Scholar + OpenAlex
  • S2被引：0
  • OpenAlex被引：0
  • 影响力被引：0
  • TLDR：Vision language models (VLMs) have achieved strong performance on visual document understanding benchmarks such as DocVQA, ChartQA, and MMLongBench-Doc. However, real-world documents combine multiple factors such as length, layout complexity, modality, and question difficulty, which makes it difficult to attribute model failures to specific causes. We introduce SynthDocBench, a fully synthetic benchmark for long-context visual document understanding that systematically controls factors including document length, layout structure, modality composition, and question type. The benchmark is constr
  • OpenAlex ID：W7168273980
  • OpenAlex DOI：10.48550/arxiv.2607.10400
  • DOI：10.48550/arxiv.2607.10400
  • DOI来源：OpenAlex
  • 开放获取：green
  • 开放获取链接：https://doi.org/10.48550/arxiv.2607.10400
  • OpenAlex更新：2026-07-19
  • 待LLM分类：否
  • 标题中文：SynthDocBench：面向长上下文视觉文档理解的受控基准
  • TLDR中文：视觉语言模型（VLMs）在 DocVQA、ChartQA、MMLongBench-Doc 等视觉文档理解基准上表现强劲。但真实文档融合长度、布局复杂度、模态、问题难度等多因素，难以将模型失败归因于具体原因。我们提出 SynthDocBench，一个全合成的长上下文视觉文档理解基准，系统性控制文档长度、布局结构、模态构成与问题类型等因子。该基准的构建……
  • 来源文件：
  • /inbox/tom/_candidates/2026-07-16-agent-rag-longcontext-candidates.json
  • [S2 enrich]
  • [OpenAlex backfill]