From Questions to Queries to Insights: A Schema-Grounded Agentic Workflow for Natural-Language Crash Data Analysis
- Published
- Source
- arXiv
- Paper number
- 1374
- Field
- RAG / Agents
- arXiv ID
- 2610.35054
Key points
- The authors proposed CrashDB-QA, a fully local, schema-grounded agentic workflow that enables natural-language analysis of crash databases.
- A shared knowledge base grounds query construction in the agency's data dictionary and safety-analysis conventions, while recorded intermediate outputs support inspection and auditing.
- They evaluated the workflow on a frozen snapshot of 4.4 million Wisconsin crash records using a benchmark of 300 questions and 9 local open-weight models.
- A code-specialized 7B model reached 80.7% accuracy at a median latency of 9.2 s, the best accuracy-latency balance among the tested models.
- Comparisons between groups or periods stayed hard, and no model exceeded 43.3% on them.
- When the Judge used authoritative field definitions and the Refiner received execution evidence, accuracy rose to 83.7%.
Paper links
External research summaries. These are not HDATF publications or measured product results.