SharedRequest: Privacy-Preserving Model-Agnostic Inference for Large Language Models 事件

SharedRequest: Privacy-Preserving Model-Agnostic Inference for Large Language Models arXiv:2606.05004v1 Announce Type: cross Abstract: With the widespread deployment of public large language models (LLMs) such as ChatGPT, protecting user prompt privacy has become an increasingly critical issue. Existing privacy-preserving inference methods sacrifice either utility or efficiency, and often require model-specific modifications that limit their compatibility. In this paper, we propose SharedReques