Researcher audits 1B LLM tokens, exposes token compression billing discrepancy

DuMate-DeepResearch system tackles auditable multi-agent reasoning for comple...

HQ arXiv: AI + Accounting arXiv · 20d ago · 1 min read

New multi-agent framework DuMate-DeepResearch addresses hallucination and verification challenges in deep research tasks through recursive search and rubric-grounded reasoning, with direct applicat...

New benchmark tests AI agents on auditable financial research workflows

HQ arXiv: AI + Accounting arXiv · 23d ago · 1 min read

Researchers introduce BigFinanceBench, a 928-item benchmark designed to evaluate financial-research AI agents on auditability and workflow transparency—measuring not just final answers but the deri...

Researchers expose LLM token billing auditing gaps as providers hide pricing ...

HQ arXiv: AI + Accounting arXiv · 28d ago · 1 min read

Academic research reveals that per-token LLM billing lacks auditability safeguards, enabling dishonest providers to overcharge—a critical issue for accounting firms using AI tools at scale.

Ember tracks AI model accuracy with year-long audited benchmark vs. predictio...

HN: Accounting & AI HN · 1mo ago · 1 min read

Ember publishes 365-day audited record comparing AI model predictions against Polymarket outcomes, scoring via Brier metric—potential framework for evaluating AI reliability in high-stakes domains ...

LLM agents audit prompt specifications in multi-agent systems

HQ arXiv: AI + Accounting arXiv · 1mo ago · 1 min read

Academic case study demonstrates iterative agent-driven auditing of prompt specifications in AEGIS, a 7-lane production pipeline, applying structured inspection rigor to LLM orchestration quality a...