用图像压缩表格，节省41%令牌且准确率提升7点
A Table Is Worth 64 Tokens: Pixel-level Compression for Multi-Table Document Question Answering
- 先用像素压缩识别相关表格，再以原分辨率推理
- 相比原方法节省41%令牌，准确率提高7个百分点
- 无需训练，适合长文档多表问答场景
在包含多个表格的文档中进行问答，需处理长序列输入，其中文本与表格交错。光学上下文压缩将上下文表示为图像，有望降低令牌消耗，但其对表格理解的影响尚不明确。我们研究了像素级表格压缩在多表文档问答中的应用，在两个基准上评估了五种视觉语言模型（VLMs）及五种视觉令牌预算。以原始分辨率将表格表示为图像时，性能与文本相当；但降采样后，模型需更长的推理路径弥补可读性损失，抵消预期节省。高度降采样的表格仍保留足够信号以判断是否相关。我们利用这一不对称性，提出一种免训练两步法：首先从像素压缩上下文中识别回答问题所需的表格，再在原分辨率下推理。在长文档上，该方法总令牌数减少41%，准确率比单步原分辨率问答高7点；同时比最高效的单步压缩配置少用15%令牌，且无准确率损失。
原文摘要 · Abstract (English)
Answering questions over real-world documents requires processing long inputs that interleave text with tables. Optical context compression, which represents context as images, promises to reduce token cost, but its effect on table understanding remains unclear. We study pixel-level table compression for question answering over documents with multiple tables, evaluating five VLMs across two benchmarks and five visual-token budgets. Representing tables as images at native resolution matches text in both performance and efficiency, but downscaling them makes models compensate the loss in readability with longer, less effective reasoning traces that cancel the expected savings. Highly downscaled tables, however, preserve enough signal to identify whether they are relevant to a question. We exploit this asymmetry with a training-free, two-step method: the model first identifies the tables needed to answer a question from a pixel-compressed context, and then reasons over those at native resolution. On long documents, our method saves 41% of total tokens and gains 7 accuracy points over single-step QA with native resolution tables. It also uses 15% fewer tokens than the most efficient single-step compressed configuration, with no accuracy loss.
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。