---
title: "When Self-Verification Hurts: Calibrated Answer Repair for Financial Question Answering"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.05166v1"
permalink: "https://feed.astralor.com/items/itm_795558ae2e1b66d54457e72b"
published: "2026-11-03T16:36:53.000Z"
score: 58
channel: "tech"
tags: ["多模态","When Self-Verification Hurts","Calibrated Answer Repair","Financial Question Answering","Financial","Asking","Based"]
---

# When Self-Verification Hurts: Calibrated Answer Repair for Financial Question Answering

## At a glance

Financial question answering can fail even when a model extracts a plausible number from the supplied evidence: the answer may attach it to the wrong entity, period, unit, or accounting concept. Asking the model to veri…

## Key points

- Financial question answering can fail even when a model extracts a plausible number from the supplied evidence: the answer may attach it to the wrong entity, period, unit, or accounting concept.
- Asking the model to verify and revise its draft can help, but open-ended revision may alter content that was already correct or inspect the wrong relation and reinforce the original error.
- We formulate repair as a decision that balances the information it may recover against the correct content it may damage.

## Full text

Financial question answering can fail even when a model extracts a plausible number from the supplied evidence: the answer may attach it to the wrong entity, period, unit, or accounting concept. Asking the model to verify and revise its draft can help, but open-ended revision may alter content that was already correct or inspect the wrong relation and reinforce the original error. We formulate repair as a decision that balances the information it may recover against the correct content it may damage. Based on this trade-off, we introduce Calibrated Answer Repair (CAR), which first estimates whether a draft is likely to benefit from repair, then selects a targeted check, localizes the relevant evidence or draft span, and limits the permitted edit. Across FinRank, Fin-RATE, and FinQA with three open-weight models, CAR outperforms generic verification in all dataset-model settings. At the system level, it recovers 14.6% of initially missed answer requirements while damaging 2.1% of initially correct ones. Its post-draft token cost is over 30 percent lower than the iterative revision alternatives. Ablations on two dataset-model settings show that evidence localization and bounded editing are the main contributors to CAR's gains.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_795558ae2e1b66d54457e72b