https://openreview.net/forum?id=BlzBcWYmdB&referrer=%5Bthe%20profile%20of%20Haoxi%20Li%5D(%2Fprofile%3Fid%3D~Haoxi_Li1)
Recent studies have shown that pre-trained vision-language models can effectively adapt to diverse downstream tasks through parameter-efficient prompt tuning....
spurious correlationcrossmodalmitigationprompt