Jacob Austin

Google DeepMind researcher known for the MBPP code benchmark, program synthesis, and large-model scaling work.

Role: researcher
Currently at: Google DeepMind
Twitter: twitter.com/jacobaustin132
GitHub: github.com/jacobaustin123
Scholar: scholar.google.com/citations
Papers: 4

Cite

Notes

Only stored in your browser.

Attribution

Affiliations & profile: scholar.google.com/citations

Attribution policy →

4papers·1eval contribs

Authored papers

Measuring The Impact Of Programming Language Distribution

arXiv 2023

2023

Scaling Up Models and Data with $\texttt{t5x}$ and $\texttt{seqio}$

arXiv 2022

2022

Language Model Cascades

arXiv 2022

2022

Program Synthesis with Large Language Models

preprint

2021

Eval contributions

Mostly Basic Python Problems (MBPP)

Google Research

974 short crowd-sourced Python tasks with three unit tests each, used alongside HumanEval as a baseline code-generation benchmark.

SaturatedCode GenerationCode

Affiliations

Currently at

Google DeepMind

researcher · frontier lab

Previously

Google Brainfrontier lab

Frequent co-authors

from 4 papers

Aitor Lewkowycz

2 shared papers

Charles Sutton

researcher

2 shared papers

David Dohan

researcher

2 shared papers

Henryk Michalewski

researcher

2 shared papers

Maarten Bosma

researcher

2 shared papers

Xavier Garcia

2 shared papers

Aakanksha Chowdhery

researcher

Adam Roberts

Afroz Mohiuddin

Alex Salcianu