arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2601.05500cs.AI
收藏

人工智能专家能力的幻觉:通过概率范式导航 elusive 的真实地面真相

The Illusion of AI Expertise Under Uncertainty: A Probabilistic Paradigm to Interpret Results Under Ground Truth Variation

Independent Researcher(独立研究员);Amazon(亚马逊);Gazi University(加齐大学);University of Colorado(科罗拉多大学);Samsun University(萨姆松大学);The University Of Melbourne(墨尔本大学);New York University(纽约大学);RMIT University(皇家墨尔本理工大学);University of Pennsylvania(宾夕法尼亚大学)

Aparna Elangovan, Lei Xu, Mahsa Elyasi, Ismail Akdulum, Mehmet Aksakal, Enes Gurun, Brian Hur, Saab Mansour, Ravid Shwartz Ziv, Karin Verspoor, Dan Roth

AI总结:

本文提出概率范式,揭示高置信度真实答案对专家高分的必要性,指出在真实答案变化大的数据集中,随机标注者与专家表现差异不大,强调在低专家表现时,按真实答案概率分层评估至关重要。

摘要:

评估AI系统的能力,包括大型语言模型和视觉模型,通常忽略了底层专家真实答案中的不确定性影响。这种模糊性不仅限于人类偏好,甚至在医疗等安全关键领域也普遍存在。本文引入概率范式,理论解释高置信度真实答案对专家高分的必要性,指出在真实答案变化大的数据集中,随机标注者与专家表现差异不大。这种特性也体现在模型比较中,不确定性会模糊表现差异。因此,忽略真实答案不确定性可能导致错误结论,即非专家与专家表现相似。通过概率范式,本文提出预期准确率和预期F1,并比较专家人类或系统在6个数据集和9个模型中,给定真实答案变化下的估计得分。结果表明,当专家表现较低时,按真实答案概率分层评估至关重要。在分层评估中,高置信度分组中的性能比较更可靠,缓解了关键混淆因素——不确定性的影响。

英文摘要:

Benchmarking the capabilities of AI systems, including Large Language Models (LLMs) and Vision Models, typically ignores the impact of uncertainty in the underlying ground truth answers from experts. This ambiguity is not just limited to human preferences, but is also consequential even in safety critical domains such as medicine where uncertainty is pervasive. In this paper, we introduce a probabilistic paradigm to theoretically explain how high certainty in ground truth answers is almost always necessary for even an expert to achieve high scores. In contrast, in datasets with high variation in ground truth answers, the expert performance drops substantially resulting in almost no difference between a random labeller and an expert. This characteristic also manifests when comparing models, where uncertainty obfuscates differences between poor and high performing models. Therefore, ignoring uncertainty in ground truth evaluation data can result in the misleading conclusion that a non-expert has similar performance to that of an expert. Given ground truth answer variability and commonly used F1 and accuracy metrics, we bring forth a simple and explainable probabilistic coin toss model to estimate expected accuracyand expected F1 and compare the estimated scores with those that experts (human or model) achieve across multiple domains -- including radiology and models such as GPT-5 and Gemini-3. The results lead to the recommendation that stratification by the probability of the ground truth answer becomes critical when expert performance is relatively low. Under stratified evaluation, performance comparison becomes more reliable in high certainty bins, mitigating the effect of the key confounding factor --- uncertainty.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑