4 sources
A high-signal researcher for the latency and systems side of modern language models, especially where clever decoding tricks turn frontier models into usable products.
Start HereYaniv Leviathan
Researcher Profile
Editor reviewedMatan Kalman
Faster LLM inference via speculative decoding
Researcher at Google Research working on faster inference and transformer efficiency
An important systems page because he is one of the named authors on speculative decoding, a technique that became part of the mainstream conversation about making large-model inference materially faster without changing outputs.
Organizations
Google Research
About This Page
This profile is meant to help you get oriented quickly: why this researcher matters, what to read first, and where to explore next.
Last reviewed
March 18, 2026
Known For
The ideas, systems, and research directions that make this person worth knowing.
01
Speculative decoding
02
Faster transformer inference
03
Efficiency-oriented model design
04
Faster LLM inference via speculative decoding
05
Fast Inference from Transformers via Speculative Decoding
06
Inference
Start Here
Canonical papers, project pages, or repositories that anchor this profile.
Signature Works
Additional papers, projects, or repositories that help flesh out the profile.
Supporting Sources
Additional links that help verify and flesh out this profile.
Related Researchers
People worth exploring next because they share topics, labs, or source material with this profile.
3 sources
Important because his profile sits at the intersection of field-level research leadership and concrete systems work such as speculative decoding that directly changed how modern LLM inference gets deployed.
Start HereYossi Matias
4 sources
A useful person to study if you care about alignment proposals that try to make superhuman systems legible enough for humans to supervise in practice.
4 sources
A high-signal researcher for understanding how DeepMind approaches generality, especially in areas where reinforcement learning, multimodality, and large-scale systems meet.
4 sources
Important for understanding how multilingual NLP, translation, and multimodal reasoning meet inside production-scale frontier systems rather than staying separate research tracks.
Start HereRadu Soricut
4 sources
A high-signal reinforcement-learning researcher whose work sits on the path from AlphaGo-era planning systems to Gemini-era reasoning and post-training techniques.