R2-D2: A Modular Baseline for Open-Domain Question Answering
This work addresses the problem of improving accuracy in open-domain QA for AI systems, presenting a modular approach that is incremental over existing methods.
The authors tackled open-domain question answering by introducing R2-D2, a four-stage pipeline combining retrieval, reranking, extractive reading, and generative reading, which achieved state-of-the-art results on NaturalQuestions and TriviaQA datasets with improvements up to 5 exact match points.
This work presents a novel four-stage open-domain QA pipeline R2-D2 (Rank twice, reaD twice). The pipeline is composed of a retriever, passage reranker, extractive reader, generative reader and a mechanism that aggregates the final prediction from all system's components. We demonstrate its strength across three open-domain QA datasets: NaturalQuestions, TriviaQA and EfficientQA, surpassing state-of-the-art on the first two. Our analysis demonstrates that: (i) combining extractive and generative reader yields absolute improvements up to 5 exact match and it is at least twice as effective as the posterior averaging ensemble of the same models with different parameters, (ii) the extractive reader with fewer parameters can match the performance of the generative reader on extractive QA datasets.