2025/10/02 by Shicheng Liu, Kai Sun, Liu, Shicheng +18
Computer Science · #Computation and Language (cs.CL) #Disk formatting #FOS: Computer and information sciences #Fuzzy Logic and Control Systems #Generalization #Information extraction #Natural Language Processing Techniques #Reinforcement learning #Scalability #Scripting language #Similarity (geometry) #Time Series Analysis and Forecasting #Web page
paper · pdf · doi:10.48550/arxiv.2510.01832
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2025/10/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Semi-structured content in HTML tables, lists, and infoboxes accounts for a substantial share of factual data on the web, yet the formatting complicates usage, and reliably extracting structured information from them remains challenging. Existing methods either lack generalization or are resource-intensive due to per-page LLM inference. In this paper, we introduce SCRIBES (SCRIpt-Based Semi-Structured Content Extraction at Web-Scale), a novel reinforcement learning framework that leverages layout similarity across webpages within the same site as a reward signal. Instead of processing each page individually, SCRIBES generates reusable extraction scripts that can be applied to groups of structurally similar webpages. Our approach further improves by iteratively training on synthetic annotations from in-the-wild CommonCrawl data. Experiments show that our approach outperforms strong baselines by over 13% in script quality and boosts downstream question answering accuracy by more than 4% for GPT-4o, enabling scalable and resource-efficient web information extraction.