Enhancing Robustness of Retrieval-Augmented Language Models with In-Context Learning

TL;DR

This paper proposes an in-context learning method to improve the robustness of Retrieval-Augmented Language Models in open-domain question answering, especially in handling unanswerable and conflicting queries without extra fine-tuning.

Contribution

The study introduces a novel in-context learning approach using MRC demonstrations to enhance RALMs' reasoning in imperfect retrieval scenarios.

Findings

Increases accuracy in identifying unanswerable questions.

Improves detection of conflicting information.

Works without additional fine-tuning.

Abstract

Retrieval-Augmented Language Models (RALMs) have significantly improved performance in open-domain question answering (QA) by leveraging external knowledge. However, RALMs still struggle with unanswerable queries, where the retrieved contexts do not contain the correct answer, and with conflicting information, where different sources provide contradictory answers due to imperfect retrieval. This study introduces an in-context learning-based approach to enhance the reasoning capabilities of RALMs, making them more robust in imperfect retrieval scenarios. Our method incorporates Machine Reading Comprehension (MRC) demonstrations, referred to as cases, to boost the model's capabilities to identify unanswerabilities and conflicts among the retrieved contexts. Experiments on two open-domain QA datasets show that our approach increases accuracy in identifying unanswerable and conflicting scenarios without requiring additional fine-tuning. This work demonstrates that in-context learning can effectively enhance the robustness of RALMs in open-domain QA tasks.