Trusted Weights, Treacherous Optimizations? Optimization-Triggered Backdoor Attacks on LLMs

Excerpt

arXiv:2605.20641v2 Announce Type: replace-cross Abstract: Inference optimization aims to minimize the latency and resource consumption of LLM inference while preserving output quality, making large-scale deployment practical and cost-effective. However, optimized execution can introduce small numerical inconsistencies from the original model. We reveal that this inconsistency not only causes the model's outputs to diverge, but more critically can introduce hidden backdoors. The backdoor remains