Rethinking Label Smoothing on Multi-hop Question Answering

TL;DR

This paper introduces a novel label smoothing technique called F1 Smoothing and a Linear Decay Label Smoothing Algorithm to improve multi-hop question answering systems, leading to better generalization and state-of-the-art results.

Contribution

It proposes a new label smoothing method tailored for MRC tasks and a curriculum-inspired decay algorithm to enhance multi-hop QA performance.

Findings

Achieved new state-of-the-art on HotpotQA leaderboard.

Enhanced generalization and robustness of multi-hop QA models.

Demonstrated effectiveness of label smoothing in multi-hop reasoning.

Abstract

Multi-Hop Question Answering (MHQA) is a significant area in question answering, requiring multiple reasoning components, including document retrieval, supporting sentence prediction, and answer span extraction. In this work, we analyze the primary factors limiting the performance of multi-hop reasoning and introduce label smoothing into the MHQA task. This is aimed at enhancing the generalization capabilities of MHQA systems and mitigating overfitting of answer spans and reasoning paths in training set. We propose a novel label smoothing technique, F1 Smoothing, which incorporates uncertainty into the learning process and is specifically tailored for Machine Reading Comprehension (MRC) tasks. Inspired by the principles of curriculum learning, we introduce the Linear Decay Label Smoothing Algorithm (LDLA), which progressively reduces uncertainty throughout the training process. Experiment on the HotpotQA dataset demonstrates the effectiveness of our methods in enhancing performance and generalizability in multi-hop reasoning, achieving new state-of-the-art results on the leaderboard.