ZJULiHongxin

I have downloaded the preprocessed R2R datasets from [this official website](https://jacobkrantz.github.io/vlnce/data). In {split}_gt.json.gz, the field 'actions' contains ground truth actions, which should produce the coordinates stored in the field 'locations'....

How to continue fine-tuning the SeeClick with Lora on Mind2WEb/AITW data?

1

Thanks for the great work! @njucckevin I tried reproducing SeeClick's performances on AITW and Mind2WEb but encountered a problem. After finetuing Qwen-VL with the 1M data mentioned in your paper,...

Any suggestions about parsing the Human Trajectories to get (s_t, a_t, s_t+1) triplets?

2

Hello! @shuyanzhou Thanks for sharing the human trajectories. I wonder if there is a way to accurately parse the human trajectories into ```(s_t, a_t, s_t+1)``` triplets. I tried to parse...