大模型
arxiv.org ·
答案正确，推理错误：准确率、一致性和共识是临床决策支持中 LLM 忠实度的误导性指标
该 arXiv 论文提出三个轻量可解释指标衡量临床 LLM 的忠实度缺口：Explanation Stability Index（ESI，重复查询下的推理一致性）、Causal Faithfulness Score（CFS，通过概念消融检验被引用概念是否真正驱动预测）、Perturbation Stability Score（PSS，语义保持改写下的鲁棒性）。在 6 个 LLM、150 道 MedQA-USMLE 题（900 次模型-问题观测）上，仅 23.3% 被引用临床概念具有因果必要性；正确答案的 CFS 低于错误答案（0.212 vs 0.398）；答案一致性与 CFS 负相关（Spearman r=-0.466）；模型对之间答案一致时仅共享 8.8% 的推理概念。作者指出准确率、一致性和共识是不完整的临床决策支持安全信号，并说明证据为行为层面（概念消融测试输出的反事实敏感性），而非内部机制。
AI推荐理由 arXiv 论文，提出 ESI、CFS、PSS 三个可解释指标，在 150 道 MedQA-USMLE 题、6 个 LLM 共 900 次观测上量化临床 LLM 的忠实度缺口：仅 23.3% 被引用临床概念具有因果必要性，正确答案的 CFS 反而低于错误答案（0.212 vs 0.398），答案一致性与 CFS 负相关（Spearman r=-0.466），模型间答案一致时仅共享 8.8% 的推理概念。作者明确说明证据是行为层面的概念消融反事实敏感性，而非内部机制，因此结论限于临床决策支持场景，不能外推为通用 LLM 忠实度结论。
78
正文
Computer Science > Artificial Intelligence
arXiv:2609.32817 (cs)
[Submitted on 26 Sep 2026]
Title:Right Answer, Wrong Reason: Accuracy, Consistency, and Consensus Are Misleading Indicators of LLM Faithfulness in Clinical Decision Support
Authors:Bharath Kumar Bolla, Bharath Kumar Bolla, Vishnu Surya Reddy Nandi
View a PDF of the paper titled Right Answer, Wrong Reason: Accuracy, Consistency, and Consensus Are Misleading Indicators of LLM Faithfulness in Clinical Decision Support, by Bharath Kumar Bolla and 1 other authors
View PDF
Abstract:Clinical Large Language Models (LLMs) achieve strong medical-exam accuracy; however, a correct answer does not guarantee that the explanation names the concepts that actually drove the decision. We introduce three lightweight, directly interpretable metrics for this faithfulness gap: the Explanation Stability Index (ESI), which measures reasoning consistency across repeated queries; the Causal Faithfulness Score (CFS), which tests whether cited concepts drive predictions via concept ablation; and the Perturbation Stability Score (PSS), which measures robustness to semantic-preserving paraphrases. By evaluating six LLMs on 150 MedQA-USMLE questions (900 model-question observations), we found that only 23.3% of the cited clinical concepts were causally necessary. Correct answers had lower CFS than incorrect answers (0.212 vs. 0.398), answer consistency negatively predicted CFS (Spearman r = -0.466), and model pairs could agree on answers while sharing only 8.8% of cited reasoning concepts. These results show that accuracy, consistency, and consensus are incomplete safety signals for clinical decision-making support. The evidence is behavioral rather than mechanistic: concept ablation tests counterfactual sensitivity of outputs, not internal circuits.
Comments:
Subjects:
Artificial Intelligence (cs.AI)
Cite as:
arXiv:2609.32817 [cs.AI]
(or
arXiv:2609.32817v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.32817
Focus to learn more
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Vishnu Surya Reddy Nandi [view email]
[v1]
Sat, 26 Sep 2026 17:44:22 UTC (619 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Right Answer, Wrong Reason: Accuracy, Consistency, and Consensus Are Misleading Indicators of LLM Faithfulness in Clinical Decision Support, by Bharath Kumar Bolla and 1 other authors
View PDF
view license
Current browse context:
cs.AI
new
|
recent
| 2026-09
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
export BibTeX citation
Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? |
Disable MathJax (What is MathJax?)