Allegro applies multi-stage programming to PBT generators and pairs it with faster randomness to achieve up to 13x faster bug discovery while exactly preserving generator semantics.
Krishnaswami and Jeremy Yallop
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
EncFormer reduces online MPC communication by 1.4x-30.4x and end-to-end latency by 1.3x-9.8x versus prior hybrid FHE-MPC systems for private GPT- and BERT-style inference while preserving accuracy.
AEGIS reduces inter-GPU communication by up to 81.3% in self-attention and reaches 96.62% scaling efficiency with 3.86x speedup on four GPUs for 2048-token encrypted Transformer inference.
citing papers explorer
-
Fail Faster: Staging and Fast Randomness for High-Performance PBT
Allegro applies multi-stage programming to PBT generators and pairs it with faster randomness to achieve up to 13x faster bug discovery while exactly preserving generator semantics.
-
EncFormer: Secure and Efficient Transformer Inference over Encrypted Data
EncFormer reduces online MPC communication by 1.4x-30.4x and end-to-end latency by 1.3x-9.8x versus prior hybrid FHE-MPC systems for private GPT- and BERT-style inference while preserving accuracy.
-
AEGIS: Scaling Long-Sequence Homomorphic Encrypted Transformer Inference via Hybrid Parallelism on Multi-GPU Systems
AEGIS reduces inter-GPU communication by up to 81.3% in self-attention and reaches 96.62% scaling efficiency with 3.86x speedup on four GPUs for 2048-token encrypted Transformer inference.
- Bonsai: Compiling Queries to Pruned Tree Traversals