VQA Counterexamples
This dataset provides a new split of VQA v2 (similarly to VQA-CP v2), which is built of questions that are hard to answer for biased models.
This dataset is designed to penalize biases, and encourage the learning of models that generalize well.
Variants: VQA-CE
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Visual Question Answering (VQA) | LMH + RMFE | Beyond Question-Based Biases: Assessing Multimodal … | 2021-04-07 |
| Visual Question Answering (VQA) | BLOCK | Beyond Question-Based Biases: Assessing Multimodal … | 2021-04-07 |
| Visual Question Answering (VQA) | RUBi | Beyond Question-Based Biases: Assessing Multimodal … | 2021-04-07 |
| Visual Question Answering (VQA) | RandImg | Beyond Question-Based Biases: Assessing Multimodal … | 2021-04-07 |
| Visual Question Answering (VQA) | LMH + CSS | Beyond Question-Based Biases: Assessing Multimodal … | 2021-04-07 |
| Visual Question Answering (VQA) | LFF | Beyond Question-Based Biases: Assessing Multimodal … | 2021-04-07 |
| Visual Question Answering (VQA) | LMH | Beyond Question-Based Biases: Assessing Multimodal … | 2021-04-07 |
| Visual Question Answering (VQA) | UpDown | Beyond Question-Based Biases: Assessing Multimodal … | 2021-04-07 |
| Visual Question Answering (VQA) | ESR | Beyond Question-Based Biases: Assessing Multimodal … | 2021-04-07 |
Recent papers with results on this dataset: