Most Cited 2025 Spotlight by Jonathan Mamou Papers
2 papers found
Conference
#1
Accelerating LLM Inference with Lossless Speculative Decoding Algorithms for Heterogeneous Vocabularies
Nadav Timor, Jonathan Mamou, Daniel Korat et al.
ICML 2025oralarXiv:2502.05202
9
citations
#2
Distributed Speculative Inference (DSI): Speculation Parallelism for Provably Faster Lossless Language Model Inference
Nadav Timor, Jonathan Mamou, Daniel Korat et al.
ICLR 2025posterarXiv:2405.14105
7
citations