Zero-Shot Estimation of Base Models' Weights in Ensemble of Machine Reading Comprehension Systems for Robust Generalization
This addresses the challenge of making MRC models more applicable to real-world general-purpose question answering by enhancing their robustness to domain shifts, though it appears incremental as it builds on ensemble techniques.
The paper tackles the problem of fragile out-of-domain generalization in machine reading comprehension models by proposing a zero-shot weighted ensemble method, which improves accuracy and robustness against domain changes.
One of the main challenges of the machine reading comprehension (MRC) models is their fragile out-of-domain generalization, which makes these models not properly applicable to real-world general-purpose question answering problems. In this paper, we leverage a zero-shot weighted ensemble method for improving the robustness of out-of-domain generalization in MRC models. In the proposed method, a weight estimation module is used to estimate out-of-domain weights, and an ensemble module aggregate several base models' predictions based on their weights. The experiments indicate that the proposed method not only improves the final accuracy, but also is robust against domain changes.