search

PhoebusSi / VQA-VS

Code for our EMNLP-2022 paper: "Language Prior Is Not the Only Shortcut: A Benchmark for Shortcut Learning in VQA"
35 stars 2 forks source link

Code #1

Closed darwann closed 1 year ago

darwann commented 1 year ago

Thank you for your great work! I'd like to replicate the experiments on my local machine; would you mind sharing the codes for all of the base models in Table 1? i.e., S-MRL, UpDn, and BAN.

PhoebusSi commented 1 year ago

Hi! Sorry for my late response. The codes for LMH and UpDn are modified from [https://github.com/chrisc36/bottom-up-attention-vqa], codes for S-MRL and RUBi are modified from [https://github.com/cdancette/rubi.bootstrap.pytorch], codes for BAN and LPF are modified from [https://github.com/jokieleung/LPF-VQA].