X

Vous n'êtes pas connecté

Maroc Maroc - TECHXPLORE.COM - RSS news feed - Hier 16:10

'Neuron-freezing' technique can stop LLMs from giving users unsafe responses

Researchers have identified key components in large language models (LLMs) that play a critical role in ensuring these AI systems provide safe responses to user queries. The researchers used these insights to develop and demonstrate AI training techniques that improve LLM safety while minimizing the "alignment tax," meaning the AI becomes safer without significantly affecting performance.

Articles similaires

Sorry! Image not available at this time

SoulMate LLM accelerator evolves according to the specific characteristics of the user

techxplore.com - 18/Mar 13:40

While large language models (LLMs) like ChatGPT are adept at answering countless questions, they often remain unaware of a user's minor habits or...

Sorry! Image not available at this time

A better method for identifying overconfident large language models

techxplore.com - 19/Mar 13:00

Large language models (LLMs) can generate credible but inaccurate responses, so researchers have developed uncertainty quantification methods to check...

Sorry! Image not available at this time

A better method for identifying overconfident large language models

techxplore.com - 19/Mar 13:00

Large language models (LLMs) can generate credible but inaccurate responses, so researchers have developed uncertainty quantification methods to check...

Sorry! Image not available at this time

Top AI coding tools make mistakes one in four times, study shows

techxplore.com - 17/Mar 15:20

New research from the University of Waterloo shows that artificial intelligence (AI) still struggles with some basic software development tasks,...

Sorry! Image not available at this time

Top AI coding tools make mistakes one in four times, study shows

techxplore.com - 17/Mar 15:20

New research from the University of Waterloo shows that artificial intelligence (AI) still struggles with some basic software development tasks,...

Sorry! Image not available at this time

New 'renewable' benchmark streamlines LLM jailbreak safety tests with minimal human effort

techxplore.com - 11/Mar 19:00

As new large language models, or LLMs, are rapidly developed and deployed, existing methods for evaluating their safety and discovering potential...

Sorry! Image not available at this time

South African Universities Join Forces to Build AI Language Models for isiXhosa, isiZulu and Sepedi

iafrica.com - 12:46

A group of South African researchers is working to ensure that African languages are not left behind as artificial intelligence reshapes how people...

Sorry! Image not available at this time

Highly performing AI agents can still fail to spot deception, study finds

techxplore.com - 21/Mar 15:00

Large language models (LLMs), artificial intelligence systems that can process and generate texts in different languages, are now used daily by many...

Sorry! Image not available at this time

Highly performing AI agents can still fail to spot deception, study finds

techxplore.com - 21/Mar 15:00

Large language models (LLMs), artificial intelligence systems that can process and generate texts in different languages, are now used daily by many...

Google launches free AI and cloud courses: Learn LLMs, image generation, and cloud engineering

times of india - 19/Mar 07:53

Google has launched free online courses in AI, machine learning, and cloud computing, targeting beginners and professionals. These modules cover...