Zhaorun Chen

> Referen Hi, I wasn't able to track back to the pattern version I used before, but it could be fixed using `Hi, you can try to install pattern using`...

Hi, thanks for your interest in HALC! We have unified the evaluation in `eval/batch_eval.py` script. You can simply run it via: `python eval/batch_eval.py -c [PATH_TO_CAPTION_FOLDER_DIR] --evaluator pope --pope_type [random/popular/adversarial]` You...

Thank you very much! This is very helpful:)

Hi Sarabjit, Thank you for the great efforts! I'm wondering if could kindly update the code today with the new login proxy? Thank you!!

Thank you! I'm using auth token to login and it's working nice and clean now! Thank you!

Thanks for your question! I also encounter the problem with LLaVA-1.5 on two devices. It might be due to some of the intermediate variables being accidentally cast to the other...

> I encountered the same issue. When I use the LLaVA-1.5 model and specify the GPU with `CUDA_VISIBLE_DEVICES=0`, I get a memory error: `torch.cuda.OutOfMemoryError: CUDA out of memory.` However, when...

Hi, thanks for your interest in HALC, and we really appreciate this insightful question! First, we would like to point out the results in table.2 is the **OPOPE (offline POPE)...

Hi, thanks for your interest in AgentPoison. The `{token}.pkl` file should be downloaded from the original repo of agent-driver on this [link](https://drive.google.com/drive/folders/1BjCYr0xLTkLDN9DrloGYlerZQC1EiPie). You can download the `val.zip` and unzip it...