BRA-Audit: Budgeted Runtime Auditing for LLM Multi-Agent Systems via Cumulative-Exposure Audit-Point Placement
3.60T1 sourcearXiv cs.MA
Source record
Published by arXiv cs.MA (T1 source). The original is at https://arxiv.org/abs/2608.14668.
Pipeline notes
The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.
SummaryBRA-Audit is a budget-aware runtime auditing framework for LLM multi-agent systems that models execution as a dynamic dependency graph. A greedy scheduler places audit points to minimize unchecked exposure, reducing token consumption by 17.2–40.6% while preserving detection performance near full-audit baselines.
Why it mattersConcrete framework with measured token savings for adding safety auditing to multi-agent LLM workflows without losing guard effectiveness.
Cited by
No citations on record.
