FinVeriQA: Financial Research Agents for Question Answering: Routing, Retrieval, MCP Tool Use, and Provenance-Aware Evidence Verification
4.60T1 sourcearXiv cs.MA
Source record
Published by arXiv cs.MA (T1 source). The original is at https://arxiv.org/abs/2611.25071.
Pipeline notes
The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.
SummaryFinVeriQA is an open-source multi-agent financial research prototype that routes questions with a small typed model, computes figures in Python, retrieves SEC filings and market data, generates cited answers, and validates references through an MCP server. Experiments compare routing and verification methods and identify failures from corrupted shared evidence.
Why it mattersMeasured failure cases and comparison data make this useful for designing agent boundaries and audit checks in evidence-sensitive research workflows.
Cited by
No citations on record.