TraVLR: Now You See It, Now You Don't! A Bimodal Dataset for Evaluating Visio-Linguistic Reasoning

1 code implementation • 21 Nov 2021 • Keng Ji Chow, Samson Tan, Min-Yen Kan

Furthermore, existing V+L benchmarks often report global accuracy scores on the entire dataset, making it difficult to pinpoint the specific reasoning tasks that models fail and succeed at.

Representation Learning