When Meaning Stays the Same, but Models Drift: Evaluating Quality of Service under Token-Level Behavioral Instability in LLMs
机构 * Electrical and Computer Engineering, Vanderbilt University(维珍尼亚大学电气与计算机工程系) ; Radiology and Radiological Sciences, Vanderbilt University(维珍尼亚大学放射学与放射科学系)
Comments This paper was developed for presentation at ICML 2025 Tokshop Workshop, but is now submitted as a standalone contribution