(Chat)GPT performs worse than BERT in detecting short-term semantic changes but slightly worse in detecting long-term changes.
(Chat)GPT performs worse than BERT in detecting short-term semantic changes but slightly lower in long-term changes.
(Chat)GPT performs worse than BERT in detecting short-term semantic changes but slightly lower in long-term changes.
The author compares the performance of (Chat)GPT and BERT in detecting semantic changes over time, highlighting the strengths and weaknesses of each model.