Video Language Models Struggle With Event Bookkeeping in Controlled Benchmarks

Researchers propose LEDGER framework to audit LLM agent workflows

HQ arXiv: AI + Accounting arXiv · 5d ago · 1 min read

LEDGER introduces claim-to-evidence trace graphs to make LLM agent outputs auditable, addressing the challenge of verifying correctness in autonomous workflows that execute code, edit files, and ge...

LLMs lower audit standards when primed with repair context, study finds

HQ arXiv: AI + Accounting arXiv · 7d ago · 1 min read

Research on ProcessBench traces reveals LLMs soften audit verification thresholds by 2.8–11.5 percentage points when prior repair episodes are in context, reducing false alarms across 15 model vari...

Benchmark reveals agentic AI models fail repeatedly at accounting tasks

HQ arXiv: AI + Accounting arXiv · 8d ago · 1 min read

AgentRelBench, a new reliability framework, tested nine LLM models on accounting operations and found systematic failures in irreversible tasks—exposing that single-run audits miss critical agent d...

Provenance tracking framework ensures AI agents meet audit requirements

arXiv: AI + Accounting arXiv · 10d ago · 1 min read

Research proposes Matrix, a causal-state layer for agentic workflows that enforces governed execution with inspectable provenance—critical for institutional settings like accounting where correct o...

Study proposes verification-cost metric to evaluate AI reliability in accounting

HQ arXiv: AI + Accounting arXiv · 14d ago · 1 min read

Researchers argue AI model evaluation for accounting must measure verification effort, not just correctness—identifying a new failure mode where auditors miss errors within budget constraints.