Google Research

Google's broad research organization covering AI/ML, systems, quantum, science, and product research - distinct from Google DeepMind.

Cite

Notes

Only stored in your browser.

Evals

Tools

Models

Papers

Boards

People

Evals

Evaluating models on the task of paraphrase detection by providing pairs of sentences that are either paraphrases or not.

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

23 challenging multi-step reasoning tasks distilled from BIG-Bench where prior models underperformed average humans.

Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models

TMLR

Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them

ACL

Program Synthesis with Large Language Models

preprint