TIGER delivers the first GPU-accelerated high-precision TFHE implementations for LLM nonlinear layers, with measured speedups of 7.17x for GELU, 16.68x for Softmax, and 17.05x for LayerNorm over CPU baselines.
Improved programmable bootstrapping with larger precision and efficient arithmetic circuits for tfhe
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
GPU Acceleration of TFHE-Based High-Precision Nonlinear Layers for Encrypted LLM Inference
TIGER delivers the first GPU-accelerated high-precision TFHE implementations for LLM nonlinear layers, with measured speedups of 7.17x for GELU, 16.68x for Softmax, and 17.05x for LayerNorm over CPU baselines.