RAVEN: In-Context Learning with Retrieval-Augmented Encoder-Decoder Language Models (2308.07922v3)

Abstract: In this paper, we investigate the in-context learning ability of retrieval-augmented encoder-decoder LLMs. We first conduct a comprehensive analysis of existing models and identify their limitations in in-context learning, primarily due to a mismatch between pretraining and inference, as well as a restricted context length. To address these issues, we propose RAVEN, a model that combines retrieval-augmented masked LLMing and prefix LLMing. We further introduce Fusion-in-Context Learning to enhance the few-shot performance by enabling the model to leverage more in-context examples without requiring additional training. Through extensive experiments, we demonstrate that our simple yet effective design significantly improves performance, achieving results comparable to the most advanced LLMs in certain scenarios, despite having substantially fewer parameters. Our work underscores the potential of retrieval-augmented encoder-decoder LLMs for in-context learning and encourages further research in this direction.