MAD (Single Model, Formerly CLIP-TD)

Reported on 2 benchmarks across 2 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing2 results

Visual Question Answering (VQA)onVCR (Q-A) test
Accuracy· 2022-04-22
79.6
best: 89.4 (GPT4RoI)
Multimodal Adaptive Distillation for Leveraging Unimodal Encoders for Vision-Language Tasks arXiv:2204.10496
Natural Language InferenceonSNLI-VE test
Accuracy· 2022-04-22
80.32
best: 91.2 (OFA)
Multimodal Adaptive Distillation for Leveraging Unimodal Encoders for Vision-Language Tasks arXiv:2204.10496