notesum.ai
Published at October 20Exploring Curriculum Learning for Vision-Language Tasks: A Study on Small-Scale Multimodal Training
cs.CL
cs.AI
Released Date: October 20, 2024
Authors: Rohan Saha1, Abrar Fahim1, Alona Fyshe1, Alex Murphy1
Aff.: 1University of Alberta

| Tasks | ||||||||
|---|---|---|---|---|---|---|---|---|
| C | T+C | C | T+C | C | T+C | C | T+C | |
| Winoground | 54.02 | 55.50 | 51.34 | 55.23 | 50.00 | 51.21 | 51.21 | 50.80 |
| VQAv2 | 41.22 | 41.72 | 42.84 | 43.98 | 41.99 | 43.00 | 35.93 | 40.85 |