arXiv:2608.08164v1 Announce Type: new Abstract: Knowledge Distillation is a widely adopted technique in the training and fine-tuning of large language models (LLMs) enabling transfer of structured information and functional behavior from a large teacher model to a smaller student model while signif

中文全文翻译正在补齐，可先查看下方英文原文。

[Submitted on 8 Aug 2026]

Title:STEMMA: An Adversarial Multi-Agent Framework for Evaluating Self-Identity Consistency in LLMs

View a PDF of the paper titled STEMMA: An Adversarial Multi-Agent Framework for Evaluating Self-Identity Consistency in LLMs, by Nuthakki Siva Gopala Krishna and 1 other authors

View PDF HTML (experimental)

Abstract:Knowledge Distillation is a widely adopted technique in the training and fine-tuning of large language models (LLMs) enabling transfer of structured information and functional behavior from a large teacher model to a smaller student model while significantly reducing computational costs. However, as the use of distillation increases in both scale and complexity it raises an important question about what kind of knowledge is really transferred from the teacher model. In this work, we argue that apart from the functional knowledge, student models also learn behavioral patterns, specifically how a model represents its own identity raising concerns about output homogeneity, model biases, and accountability. To address this challenge, we introduce STEMMA, a multi-modal and multi-agent framework in which role specific agents collaboratively probe self identification behavior in different models. We also contribute a set of adversarial prompts designed manually to evaluate identity consistency in LLMs. Our results show that to an extent most models are vulnerable to inconsistencies in self-representations.

Submission history

From: Siva Gopala Krishna Nuthakki [

view email

]

[v1]

Sat, 8 Aug 2026 14:46:22 UTC (4,649 KB)