{"ID":2838353,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.19480","arxiv_id":"2511.19480","title":"Exploiting the Experts: Unauthorized Compression in MoE-LLMs","abstract":"Mixture-of-Experts (MoE) architectures are increasingly adopted in large language models (LLMs) for their scalability and efficiency. However, their modular structure introduces a unique vulnerability: adversaries can attempt to compress or repurpose models by pruning experts and cheaply fine-tuning the remainder, effectively bypassing licensing and security constraints. In this paper, we systematically study the prunability of MoE-LLMs under task-specific usage. We first develop an expert attribution framework that identifies the subset of experts most responsible for a given task, then evaluate the performance trade-offs of pruning and re-aligning these experts using active learning-driven fine-tuning. Our findings reveal a critical knowledge loss--recovery trade-off: while certain experts can be isolated to retain task accuracy, significant degradation occurs without targeted re-alignment. Based on this analysis, we propose defense strategies that aim to make MoE models harder to compress and fine-tune without authorization, including entangled expert training and selective fine-tuning protocols that resist unauthorized adaptation. By positioning expert pruning as both a threat vector and a defense target, this work highlights the dual-use nature of MoE modularity and provides the first systematic evaluation framework for secure specialization of MoE-LLMs.","short_abstract":"Mixture-of-Experts (MoE) architectures are increasingly adopted in large language models (LLMs) for their scalability and efficiency. However, their modular structure introduces a unique vulnerability: adversaries can attempt to compress or repurpose models by pruning experts and cheaply fine-tuning the remainder, effe...","url_abs":"https://arxiv.org/abs/2511.19480","url_pdf":"https://arxiv.org/pdf/2511.19480v1","authors":"[\"Pinaki Prasad Guha Neogi\",\"Ahmad Mohammadshirazi\",\"Dheeraj Kulshrestha\",\"Rajiv Ramnath\"]","published":"2025-11-22T20:08:29Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}