DeepMind rolls out human-style encouragement to enhance AI math abilities

Share post:

Google DeepMind researchers have developed a new technique to improve the math ability of AI language models by using human-style encouragement. The technique, called Optimization by PROmpting (OPRO), uses natural language to guide the language model in problem-solving.

OPRO improves LLMs using natural language. It uses two LLMs; one to rate solutions and one to create them.

It works by the scorer LLM describing the problem, then the optimizer LLM finds a solution. The scorer LLM rates the solution, and the optimizer LLM adjusts based on feedback. This repeats until the scorer LLM is satisfied.

In one experiment, the researchers used OPRO to improve the performance of Google’s PaLM 2 language model on a dataset of grade-school math word problems. They found that the most effective prompt was “Take a deep breath and work on this problem step by step.” This prompt helped PaLM 2 achieve an accuracy score of 80.2%, compared to just 34% without any prompting.

The researchers believe that human-style encouragement works because it helps the language model to tap into its knowledge of human reasoning and problem-solving. They are hopeful that OPRO can be used to improve the performance of AI language models on a wide range of tasks, including education, translation, and code generation.

The sources for this piece include an article in ArsTechnica.

Featured Tech Jobs


Related articles

Research Raises Concerns Over AI Impact on Code Quality

Recent findings from GitClear, a developer analytics firm, indicate that the increasing reliance on AI assistance in software...

Microsoft to train 100,000 Indian developers in AI

Microsoft has launched an ambitious program called "AI Odyssey" to train 100,000 Indian developers in artificial intelligence by...

NIST issues cybersecurity guide for AI developers

Paper identifies the types of cyberattacks that can manipulate the behavior of artificial intelligen

Canada, U.S. sign international guidelines for safe AI development

Eighteen countries, including Canada, the U.S. and the U.K., today agreed on recommended guidelines to developers in their nations for the secure design, development, deployment, and operation of artificial intelligent systems. It’s the latest in a series of voluntary guardrails that nations are urging their public and private sectors to follow for overseeing AI in

Become a member

New, Relevant Tech Stories. Our article selection is done by industry professionals. Our writers summarize them to give you the key takeaways