Danny (Zhiwei) Xu

why large influence is harmful

4

does there have any evaluation for influence function values? I found the most helpful train image label in CIFAR10 is 2 while the test image label 4.

logical behind score() in bart_score.py

2

Hi, I am wondering how is the logical reasoning of this model. If my source and target are source: ["I have one sister and one brother"] target: [["I have no...

Hi, I saw that in autoregressive/train/extract_codes_t2i.py, it needs .jsonl file. Where or how to create this jsonl file? What is it? It is not mentioned in README.md at all. Thanks.