PACE: Publisher-Adaptive Content Extraction via Agentic Automation

PACE：基于智能体自动化的出版商自适应内容提取

Abstract: Web content extraction is essential for reliable LLM data pipelines, yet existing methods often struggle to jointly satisfy accuracy, scalability, and adaptability. General-purpose extractors can be applied broadly, but they are often brittle on publisher-specific layouts and richer extraction targets such as metadata, images, and tables. Direct LLM-based extraction offers greater flexibility, but incurs substantial cost and latency at scale, while manually engineered publisher-specific parsers can achieve high accuracy but require substantial human effort to build and maintain.

摘要： 网络内容提取对于构建可靠的大语言模型（LLM）数据流水线至关重要，然而现有方法往往难以同时兼顾准确性、可扩展性和适应性。通用提取器虽然应用广泛，但在处理特定出版商的布局以及元数据、图像和表格等更丰富的提取目标时，往往表现脆弱。直接基于 LLM 的提取虽然提供了更大的灵活性，但在大规模应用时会产生高昂的成本和延迟；而人工编写的特定出版商解析器虽然能实现高准确度，却需要投入大量人力进行构建和维护。

We introduce PACE, an agentic framework for learning publisher-specific extraction configurations from representative pages and user requirements. During training, PACE uses LLMs to analyze page structure and aggregate reusable extraction patterns. At inference time, the learned configurations instantiate a fixed deterministic extractor template, enabling scalable extraction without additional LLM calls.

我们引入了 PACE，这是一个通过代表性页面和用户需求来学习特定出版商提取配置的智能体框架。在训练阶段，PACE 利用 LLM 分析页面结构并聚合可复用的提取模式。在推理阶段，学习到的配置会实例化为一个固定的确定性提取器模板，从而在无需额外调用 LLM 的情况下实现可扩展的提取。

Experiments spanning article-body, metadata, and multimodal extraction show that PACE outperforms scalable non-manual baselines while approaching the quality of manually engineered publisher-specific parsers. PACE achieves stronger extraction of article text, metadata, images, and tables, demonstrating that agentic configuration learning can automate publisher-specific extraction for LLM-ready page representations beyond article text.

涵盖文章正文、元数据和多模态提取的实验表明，PACE 的表现优于可扩展的非人工基准方法，并接近人工编写的特定出版商解析器的质量。PACE 在文章文本、元数据、图像和表格的提取方面表现更强，证明了智能体配置学习能够实现针对 LLM 就绪页面表示的自动化提取，且其能力已超越了单纯的文章文本提取。