New benchmark evaluates LLM reliability for deterministic tasks like invoice extraction and transcript parsing—critical for automation workflows in accounting and bookkeeping.
Continue reading
Research on 99,952 examples shows correct domain rubrics improve LLM evaluator accuracy by 2.11 points, suggesting specialization matters more in evaluation rules than model weights.
Researchers introduce FinanceHarness, an agentic AI framework designed specifically for financial deep research that combines LLMs with layered agent architecture and point-in-time benchmarks to au...
Intuit expands QuickBooks integrations with Claude and ChatGPT to enable agentic tasks—creating invoices, checking payroll data, and accessing lending benchmarks—moving beyond read-only access to a...
Billtrust connects its AR platform to Claude and Microsoft Copilot, allowing AI agents to access structured invoice data with audit trail tracking but no raw data exposure.
Kimi K3 evaluated on TaxCalcBench for ability to file American tax returns, testing cross-border AI capability in specialized tax domain.