Manli Shu

Hi authors, Just want to follow up on this issue. Could you please provide more details of the RepSurf backbone for detection? Thanks!

Hi @cooleel, thank you for your interests in our work! We plan to make the model and dataset cards available later today (we're working on final reviews.) Will let you...

Thank y'all for you interests in our work. Update on this threads: all 4 models are now live on huggingface hub: https://huggingface.co/collections/Salesforce/xgen-mm-1-models-662971d6cecbf3a7f80ecc2e. For the two datasets, they're still undergoing internal...

Hi, sorry for the late update. Both datasets are available under the [xGen-MM collection](https://huggingface.co/collections/Salesforce/xgen-mm-1-models-and-datasets-662971d6cecbf3a7f80ecc2e) (They've been public for a couple days, sorry about the delayed update.) Thank you again for...

Hi @zenithc-git , Thank you for your interest in our work. TPT is a test-time method, so we didn't do any model training. In the paper, we mainly use OpenAI's...

Hi all, thank you all for trying out the code. Could you provide more details about the command you ran? Also, @zhaihaotian which "cross-dataset" was it? Sorry I don't have...

If this issue only happens when evaluating ImageNet-A and ImageNet-R, there might be something wrong with the label masking. These two datasets only have 200 of the 1000 ImageNet classes,...

Hi @weiyao-Wang, Thank you for your interest in our code. Yes, we substantially rewrote the dataloaders in this branch for supervised fine-tuning compared to the original open-flamingo implementation. You can...

Hi all, Apologies for the late reply. Thank you for your interests in adopting our code. During fine-tuning, our code does not distinguish image-text pairs from interleaved image-text data. The...

Hi @Gaojinpeng8 , Thank you for trying out our code. We noticed this issue with newer version of `transformers` and we're investigating it. In the meantime, could you try using...