vlf-silkie / VLFeedback

vlf-silkie/VLFeedback Issues

DPO performance on other models
Updated a month ago7
Is it normal for the loss to converge to 0.3 at the end of training?
Closed 6 months ago3
How could I specify the GPU index for the DPO training?
Closed 6 months ago1
Cannot reproduce MM-Vet score
Closed 6 months ago4
ValueError: 151859 is not in list
Updated 7 months ago4
The prompt of gpt4v.
Closed 6 months ago3
Impact of Including GPT-4V in LVLM Pool?
Updated a month ago1