arXivBoyang Zhang, Adrian Lyjak, Eli Stewart, Zhaoqi Li, Simon SuoFri, Jul 31, 2026, 10:55 AM PDT
score 15.4
New benchmark tests AI agents that pull data from messy business documents
Original: ExtractBench: A Benchmark for Schema-Guided Enterprise Document Extraction
Source: arxiv.org ↗
Writing ELI5 summary…