vivo artificial intelligence model appears in C-Eval

According to reports, vivo’s self-developed large-scale pre-trained language model “vivo_Agent_LM_7B” was unveiled in C-Eval and achieved impressive results

According to the test results, vivo_Agent_LM_7B scored 64.4 points in C-Eval, ranking fifth, and achieved high scores of 80-90 in multiple subjects in the field of social sciences. In addition, it is very close to the second-ranked GPT-4 (68.7 points) in terms of overall score

It is reported that vivo plans to integrate the vivo_Agent_LM_7B model into the voice assistant Jovi to improve the user experience

The above is the detailed content of vivo artificial intelligence model appears in C-Eval. For more information, please follow other related articles on the PHP Chinese website!