A study conducted by Stanford University reveals that the popular AI chatbot ChatGPT created by OpenAI is experiencing fluctuations in performance, particularly in solving mathematical equations. The research highlights the phenomenon of "drift" in AI language models, emphasizing the importance of monitoring and addressing these changes to ensure consistent and reliable performance.
By India Today Tech: A study conducted by Stanford University found that the popular AI chatbot ChatGPT, created by OpenAI, is getting worse at solving mathematical equations. The chatbot experienced significant performance fluctuations on certain tasks between March and June. The research compared two versions of the technology– GPT-3.5 and GPT-4, focusing on tasks such as solving math problems, answering sensitive questions, generating software code, and visual reasoning.
As per the Fortune report, the study revealed a phenomenon called "drift," where the technology's ability to perform specific tasks changed unpredictably over time. In the case of GPT-4's math problem-solving capability, its accuracy dropped drastically from 97.6 per cent in March to a mere 2.4 per cent in June. As per the findings, the GPT-3.5 model demonstrated an opposite trajectory, improving from 7.4 per cent accuracy in March to 86.8 per cent in June on the same task.
7th Edition of International Conference on Mathematics and Optimization Method
Website Link:https://maths-conferences.sciencefather.com/
Award Nomination: https://x-i.me/XU6E
Instagram: https://www.instagram.com/maths98574/
Twitter: https://twitter.com/AnisaAn63544725
Pinterest: https://in.pinterest.com/maxconference20022/
#maths #numericals #algebra #analysis #analysis #mathmatics #numericals #number #complex #graphics
No comments:
Post a Comment