FuseFSS: Efficient Secure LLM Inference with Function Secret Sharing
About
Two-server secure inference allows a client to query a hosted large language model (LLM) without revealing prompts or embeddings. Recent GPU systems based on function secret sharing (FSS) make linear layers efficient, but fixed-point nonlinearities and helper operations remain a bottleneck because each operator is typically implemented as a bespoke protocol with its own comparisons, wrap-around corrections, and preprocessing material. We present FuseFSS, a compiler that replaces per-operator protocol design with a single compilation pipeline. For each scalar fixed-point operator, a compact specification lists its interval partition, low-degree arithmetic pieces, and required predicate bits. The compiler emits two batched FSS evaluations on the public masked value: one packed comparison that returns all predicate bits, and one vector interval lookup that returns the active coefficients and constants. Compared to the current state-of-the-art FSS-based GPU secure inference, FuseFSS preserves accuracy while achieving a $1.24\times$--$1.50\times$ end-to-end speedup and reducing online communication by $9\%$--$16\%$ on BERT and GPT-style models; preprocessing is also lighter, with $14\%$--$23\%$ lower key-generation time and $20\%$--$24\%$ smaller keys.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Secure Inference | Transformer Models | Online Time (ms)42.6 | 10 | |
| Natural Language Understanding | GLUE | SST-2 Accuracy92.5 | 6 | |
| Secure LLM Inference | LLaMA 7B | -- | 3 | |
| Secure LLM Inference | Llama 8B 3.1 | -- | 3 |