Semantic Variability of Replies Across LLMs: Implications for Designing Conversation-Based Assessment

Abstract:This study examines whether LLM-generated replies remain semantically consistent when the underlying LLM changes. Using messages from real collaborative conversations, we compared the semantic similarity of generated replies across LLMs under two conditions: with and without preceding chat history. Results show that model choice and conversational context both affect response similarity and alignment with human replies. These findings indicate that prompting and conversational context alone may not be sufficient to preserve response consistency across LLMs, highlighting the need for infrastructure and design strategies that can maintain stable and comparable responses amid the rapid and continuous evolution of LLMs.

9 pages, 4 figures, two tables. Accepted to the AI in Measurement and Education Conference 2026

Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)

arXiv:2608.24920 [cs.CL]

(or arXiv:2608.24920v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2608.24920 Focus to learn more arXiv-issued DOI via DataCite

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.