Lianghui Zhu

Questions about Multi-view Semantic Fusion

1

Hi, I am your sincere follower! Your semantic-NeRF is a very suprising work！ I wonder if you could release the Multi-view Semantic Fusion code?

Greetings, Thank you for your code and bravo paper, they give us a novel way to process the relationship with most descriptive part and the other object part. When I...

Question about the training process

1

Greetings! Thanks for your excellent work! When running your code, I met a problem that the performance is poor. My running command is ``` bash train_val_cub.sh 3 deit small 224...

Questions about Deeplab retrain part

2

Greetings！ I read your paper again and again recently, and I really learned a lot from your work. For the retrain part, I met some problems, would you like to...

OOM error and fsdp support

5

Thanks for your excellent work! I have tried vicuna-13b on 8xA100(40G), but it will result in an OOM error. To avoid the GPU OOM error, I add the following commands...

Question about abnormal response

2

Thanks for your awesome work again! When I run the pretrained code in the latest FSDP style code, I use the command here for a test. ``` python -m llava.eval.run_llava...

What's the differences between RAFT and InstructGPT's PPO?

1

Greetings! I am very interested in your work RAFT. When I read the paper, you still use the RL methods to score the model's output. It isn't very clear to...

[Question] Can not reproduce the Acc. on Science QA datasets

4

### Question Thanks for running the script for SQA finetuning. After the fine-tuning of SQA. I find the result for the SQA test set (Total: 4241, Correct: 3670, Accuracy: 86.54%)...

Details for finetune SQA dataset

1

Greetings! I have checked the great performance on your COCO VQA datasets. Now I want to check the performance on the SQA dataset, could you give me more guidance about...

The Accuracy Problem of Mamba Operator

4

Greetings! Thanks for your awesome work! The GPU-level optimization of the mamba operator is impressive to me. But I face the accuracy problem when trying the unit test in running...