Visual Question Answering On Vqa Cp
평가 지표
Score
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | Score | Paper Title | Repository |
|---|---|---|---|
| HAN | 28.65 | Learning Visual Question Answering by Bootstrapping Hard Attention | - |
| Learned-Mixin +H | 52.05 | Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases | |
| CSS | 58.95 | Counterfactual Samples Synthesizing for Robust Visual Question Answering | |
| UpDn+SCR (VQA-X) | 49.45 | Self-Critical Reasoning for Robust Visual Question Answering | |
| LMH+Entropy regularization | 54.55 | Removing Bias in Multi-modal Classifiers: Regularization by Maximizing Functional Entropies | |
| LMH+Entropy regularization (Ensemble) | 56.74 | Removing Bias in Multi-modal Classifiers: Regularization by Maximizing Functional Entropies | |
| GGE-DQ | 57.32 | Greedy Gradient Ensemble for Robust Visual Question Answering | |
| RUBi | 47.11 | RUBi: Reducing Unimodal Biases in Visual Question Answering | |
| MuRel | 39.54 | MUREL: Multimodal Relational Reasoning for Visual Question Answering | |
| NSM | 45.8 | Learning by Abstraction: The Neural State Machine |
0 of 10 row(s) selected.