跳到正文
arXiv cs.AI· Cosimo Galeone, Anna Ettorre, Ginevra Mariani, Minsu Park, Giuseppe Ettorre, Daniele Ligorio·· 4 小时前AI 评分44

金融研究智能体可信吗?对 Vals Finance Agent Benchmark v2 参考智能体的漂移、诊断与可审计干预研究

Can You Trust Your Finance Agent? Drift, Diagnosis and Auditable Intervention on One Shipped Financial Research Agent

AI 导读

一项研究对 Vals Finance Agent Benchmark v2 随附的金融研究智能体未做调优即展开审计:同一配置在 18 道题上 7 次运行,149 项评分标准的通过数在 105 至 123 之间波动,账单差异最高达 17%。

正文 · 原文

View PDF HTML (experimental)

Abstract:Can you trust a finance agent you did not build? We take the reference agent shipped with the Vals Finance Agent Benchmark v2, leave it untuned, and ask four questions a leaderboard score does not answer. Is the score stable? Seven identical runs on 18 questions pass between 105 and 123 of 149 rubric criteria, and the bill varies by up to 17%. When the agent fails, who owns the fix? We classify 20 failed trajectories, from three runs of nine held-out questions, by the layer whose owner would remedy them, under a taxonomy frozen in advance: 13 belong to the agent's instructions, 3 to its tools, 4 to the model, none to its data. Can that diagnosis be trusted? Four blind readers from three model families, given only the traces, name the same fault as our LLM analyzer in 94% of cases, the rate at which they agree with one another (93%). Can a change be trusted? An optimization loop reads the traces, makes one targeted change per round and keeps it only if it clears a fixed threshold. Under a cost objective its final configuration costs 0.34x the baseline on the held-out questions (95% CI [0.22, 0.58]) at accuracy within noise. Under an accuracy objective it gains 14 points on the development questions and 1.5 on the held-out ones (CI [-5.3, +7.2]). Judged against the measured run-to-run noise, under a rule written before the test was opened, the first is a real saving and the second is not. This is a case study of one agent and nine test questions, not a general result.
Comments: 9 pages. Accepted as an oral presentation at the AI4F workshop, ICAIF 2026, Milan
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.17939 [cs.CL]
  (or arXiv:2611.17939v1 [cs.CL] for this version)
  https://doi.org/10.48550/arXiv.2611.17939

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Cosimo Galeone [view email]
[v1] Fri, 13 Nov 2026 09:22:23 UTC (42 KB)

来源:arXiv cs.AI · arxiv.org