harzva

why tokenized_prompts.argmax =49407，'<|endoftext|>'

1

Can represent global information of tokenized_prompts？why tokenized_prompts.argmax(dim=-1) '': 49407 like cls_token of transformer? Thanks

In the MOE method does expert have to learn and can the frozen model be used as an expert?like gpt3 bert

1

**Describe the question(问题描述)** Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts In the MOE method does expert have to learn and can the frozen model be used as an...

In the MOE method does expert have to learn and can the frozen model be used as an expert?

1

**Describe the question(问题描述)** Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts In the MOE method does expert have to learn and can the frozen model be used as an...

I would like to ask how long is the experimental period for coop training on imagenet?thanks you！

where is The knowledge extraction module in code？thanks！！！！

1

where is The knowledge extraction module in code？thanks！！！！What specific model are they using?

Image and text retrieval

2

I would like to quote your method of Qualitative analysis, method. image and text retrieval your work is very meaningful, however this piece of code did not find convenient open...