Retrieval Augmented Visual Question Answering with Outside Knowledge

7 Oct 2022  ·  Weizhe Lin, Bill Byrne ·

Outside-Knowledge Visual Question Answering (OK-VQA) is a challenging VQA task that requires retrieval of external knowledge to answer questions about images. Recent OK-VQA systems use Dense Passage Retrieval (DPR) to retrieve documents from external knowledge bases, such as Wikipedia, but with DPR trained separately from answer generation, introducing a potential limit on the overall system performance. Instead, we propose a joint training scheme which includes differentiable DPR integrated with answer generation so that the system can be trained in an end-to-end fashion. Our experiments show that our scheme outperforms recent OK-VQA systems with strong DPR for retrieval. We also introduce new diagnostic metrics to analyze how retrieval and generation interact. The strong retrieval ability of our model significantly reduces the number of retrieved documents needed in training, yielding significant benefits in answer quality and computation required for training.

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Retrieval OK-VQA RA-VQA Recall@5 82.84 # 2
Visual Question Answering (VQA) OK-VQA RA-VQA-FrDPR (T5-large) Accuracy 51.22 # 17
Exact Match (EM) 55.77 # 3
Recall@5 81.25 # 3
Visual Question Answering (VQA) OK-VQA RA-VQA (T5-large) Accuracy 54.48 # 14
Exact Match (EM) 59.41 # 2
Recall@5 82.84 # 2


No methods listed for this paper. Add relevant methods here