W
When AI models are past their prime.
A recent study found that if a coding problem put before ChatGPT (using GPT-3.5) existed on coding practice site LeetCode before its 2021 training data cutoff, it did a very good job generating functional solutions, writes IEEE Spectrum.
But when the problem was added after 2021, it sometimes didn’t even understand the questions and its success rate seemed to fall off a cliff, underscoring AI’s limitations without enough data.
How Good Is ChatGPT at Coding, Really?
[IEEE Spectrum]
Most Popular
Most Popular
OpenAI announces SearchGPT, its AI-powered search engine
The Google era is officially over
There is no fix for Intel’s crashing 13th and 14th Gen CPUs — any damage is permanent
Uber, Lyft, and DoorDash have finally won the Prop 22 gig worker battle
The Disney Plus, Hulu, and Max streaming bundle is now available