Discrepancy in Reproduced Results on VQAv2: 77.12 vs. Reported 77.9

<img width="703" height="163" alt="Image" src="https://github.com/user-attachments/assets/4127dda6-99e9-426d-b967-d1d1a05c7ccf" />

<img width="2838" height="1131" alt="Image" src="https://github.com/user-attachments/assets/5e828bb0-f9c2-46cc-bfae-7e1935caba1c" />

<img width="1437" height="755" alt="Image" src="https://github.com/user-attachments/assets/6a75eeeb-4deb-4fcb-9181-e7e3ca732386" />

Hello author, I have reproduced your paper, but when using your 7B+144Token model on the VQAv2 dataset, I obtained a result of 77.12, which significantly differs from the 77.9 reported in your paper. Could you please explain the reason for this discrepancy?

Navigation Menu

Discrepancy in Reproduced Results on VQAv2: 77.12 vs. Reported 77.9 #24

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

Discrepancy in Reproduced Results on VQAv2: 77.12 vs. Reported 77.9 #24

Description

Activity

Metadata

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

Issue actions