CORD

A systematic benchmark of open-source instruction-tuned LLMs for key-value pair extraction found that models including Gemma, Mistral, Qwen2.5, LLaMA 3, and DeepSeek approach supervised layout-aware s…

A new arXiv preprint (2608.14639v1) reports that per-field selective risk control for document extraction silently fails on real documents, diagnosing three failure modes across 13,859 genuine claude-…

A developer building AI agents for home service businesses discovered that critical customer information was being lost during handoffs to scheduling and CRM systems. They found the CORD protocol, an …