Signals 4 · free daily AI digest

Same-Number Citation Swaps: Stress-Testing Jev as a Financial Evidence Judge

Paper recorded by Signals 4 on 2026-10-06 in cs.CL. Abstract reproduced from arXiv; link to the original below.

Published 2026-10-06 on arXiv · recorded by Signals 4 on 2026-10-07

Category: cs.CL · 自然语言处理 · first seen 2026-10-07

Abstract

Financial reports repeat values across periods, metrics and accounting lines, allowing an LLM-generated calculation to be numerically correct while citing the wrong financial role. We evaluate what probabilistic evidence verification adds beyond number matching using Jev as a source-support verifier for GPT-4.1-mini calculation traces. A signed-number-at-pointer baseline explains most recovery ove

Read on arXiv →

#312 most recent of 660 cs.CL papers we have recorded · ↑ newer: Agreement Is Not Validity: Cross-Model LLM Consensus in Diagnosing Stu · ↓ older: Evidence-Bound Reasoning: Neuro-Semantic Verification of Biomedical AI
Cite this page: Same-Number Citation Swaps: Stress-Testing Jev as a Financial Evidence Judge: the #312 most recent of 660 cs.CL papers we have recorded (as of 2026-10-06). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/same-number-citation-swaps-stress-testing-jev-as-a-financial-evidence-judge.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.CL papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions