Google improves Bard's logic & reasoning skills

Bard now uses a new technique called "implicit code execution" to recognise computational prompts and run code in the background, the tech giant said in a blogpost on Wednesday. As a result, it can respond to string manipulation, coding questions and mathematical operations more correctly.


Tech giant Google has introduced new improvements in its artificial intelligence (AI)-chatbot Bard, including better logic and reasoning skills.

Bard now uses a new technique called "implicit code execution" to recognise computational prompts and run code in the background, the tech giant said in a blogpost on Wednesday.

Advertisement

As a result, it can respond to string manipulation, coding questions and mathematical operations more correctly.

"Our new method allows Bard to generate and execute code to boost its reasoning and math abilities. This approach takes inspiration from a well-studied dichotomy in human intelligence, notably covered in Daniel Kahneman's book 'Thinking, Fast and Slow'-- Athe separation of 'System 1' and 'System 2' thinking," it added.

Advertisement

System 1 thinking is fast, intuitive and effortless. On the other hand, System 2 thinking is slow, deliberate and effortful.

With the latest update, the company has combined the capabilities of both Large language models (LLMs) (System 1) and traditional code (System 2) to help improve the accuracy of Bard's responses.

Advertisement

"We've seen this method improve the accuracy of Bard's responses to computation-based word and math problems in our internal challenge datasets by approximately 30 per cent," the tech giant said.

Also, the company announced that Bard now offers a new export option to Google Sheets.

Advertisement

With the new feature, users can export the tables that Bard creates in its responses directly to Sheets.

Also Read| Amazon may launch ad-supported tier of Prime Video  

Advertisement

Also Read| Meta Verified badge arrives in India, starts from Rs 699

tags
Advertisement