ExtractBench: A Benchmark for Schema-Guided Enterprise Document Extraction

За 30 секунд

Что произошло

ExtractBench, a benchmark for evaluating schema-guided document extraction, was released with 4,869 pages across 370 enterprise documents spanning 8 business domains.

Почему это важно

Engineers building document processing systems need this to compare extraction accuracy, completeness, grounding quality, and cost across different AI agents and models.

На что обратить внимание

Vision language models handle short documents well but truncate long record lists; coding agents cost significantly more despite higher accuracy on length-heavy tasks.