RFiD: Towards Rational Fusion-in-Decoder for Open-Domain Question Answering
This addresses a reliability issue in ODQA systems for users needing accurate answers, though it appears incremental as it builds on the existing Fusion-in-Decoder approach.
The paper tackles the problem of Open-Domain Question Answering systems relying on spurious features rather than genuine causal relationships, introducing the RFiD model which improves Exact Match scores by up to 1.5 on Natural Questions and 0.7 on TriviaQA.
Open-Domain Question Answering (ODQA) systems necessitate a reader model capable of generating answers by simultaneously referring to multiple passages. Although representative models like Fusion-in-Decoder (FiD) have been proposed to address this challenge, these systems can inadvertently rely on spurious features instead of genuine causal relationships between the question and the passages to generate answers. To counter this problem, we introduce the Rational Fusion-in-Decoder (RFiD) model. Our model leverages the encoders of FiD to differentiate between causal relationships and spurious features, subsequently guiding the decoder to generate answers informed by this discernment. Experimental results on two ODQA datasets, Natural Questions (NQ) and TriviaQA (TQ), demonstrate that our model surpasses previous methods, achieving improvements of up to 1.5 and 0.7 in Exact Match scores on NQ, and exhibits an enhanced ability to identify causal relationships.