Atlas: Few-shot Learning with Retrieval Augmented Language Models

Abstract

Large language models have shown impressive few-shot results on a wide range of tasks. However, when knowledge is key for such results, as is the case for tasks such as question answering and fact checking, massive parameter counts to store knowledge seem to be needed. Retrieval augmented models are known to excel at knowledge intensive tasks without the need for as many parameters, but it is unclear whether they work in few-shot settings. In this work we present Atlas, a carefully designed and pre-trained retrieval augmented language model able to learn knowledge intensive tasks with very few training examples. We perform evaluations on a wide range of tasks, including MMLU, KILT and NaturalQuestions, and study the impact of the content of the document index, showing that it can easily be updated. Notably, Atlas reaches over 42% accuracy on Natural Questions using only 64 examples, outperforming a 540B parameters model by 3% despite having 50x fewer parameters.

Cited by in corpus (8)

• A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions
• Large Language Models for Information Retrieval: A Survey
• WikiChat: Stopping the Hallucination of Large Language Model Chatbots by Few-Shot Grounding on Wikipedia
• Knowing You Don't Know: Learning When to Continue Search in Multi-round RAG through Self-Practicing
• "In-Context Learning" or: How I learned to stop worrying and love "Applied Information Retrieval"
• Complex QA and language models hybrid architectures, Survey
• Efficient Tuning of Large Language Models for Knowledge-Grounded Dialogue Generation
• Derivation Prompting: A Logic-Based Method for Improving Retrieval-Augmented Generation