ELLMo: Packing- and Depth-Aware Encrypted Transformer Inference

본문 미리보기

Cloud-based Large Language Model (LLM) inference processes sensitive user inputs, yet current deployments offer limited confidentiality guarantees. Fully Homomorphic Encryption (FHE) can provide strong privacy, but it clashes with transformer architectures, where rigid ciphertext packing demands expensive rotations, and deep polynomial circuits for nonlinearities necessitate costly bootstrapping. Although recent work has reported promising speed-ups, maintaining model accuracy is a challenge. We