Categories: FAANG

FunSearch: Making new discoveries in mathematical sciences using Large Language Models

In a paper published in Nature, we introduce FunSearch, a method for searching for “functions” written in computer code, and find new solutions in mathematics and computer science. FunSearch works by pairing a pre-trained LLM, whose goal is to provide creative solutions in the form of computer code, with an automated “evaluator”, which guards against hallucinations and incorrect ideas.
AI Generated Robotic Content

Recent Posts

Update: Distilled v1.1 is live

We've pushed an LTX-2.3 update today. The Distilled model has been retrained (now v1.1) with…

22 hours ago

How to Implement Tool Calling with Gemma 4 and Python

The open-weights model ecosystem shifted recently with the release of the

22 hours ago

Structured Outputs vs. Function Calling: Which Should Your Agent Use?

Language models (LMs), at their core, are text-in and text-out systems.

22 hours ago

Cram Less to Fit More: Training Data Pruning Improves Memorization of Facts

This paper was accepted at the Workshop on Navigating and Addressing Data Problems for Foundation…

22 hours ago

How to build effective reward functions with AWS Lambda for Amazon Nova model customization

Building effective reward functions can help you customize Amazon Nova models to your specific needs,…

22 hours ago

How to find the sweet spot between cost and performance

At Google Cloud, we often see customers asking themselves: "How can we manage our generative…

22 hours ago