Izacard et al.: jointly pretrained few-shot retrieval-augmented model

What changed

Jointly pretrained a dense retriever and Fusion-in-Decoder generator for knowledge-intensive few-shot tasks, studying adaptation strategies and updating the external document index.

Scientific scope

Jointly pretrained a dense retriever and Fusion-in-Decoder generator for knowledge-intensive few-shot tasks, studying adaptation strategies and updating the external document index. Contriever, FiD and retrieval augmentation are inherited, not invented by Atlas. Comparative small-model/few-shot results apply to stated tasks and budgets; index updates do not guarantee accurate or complete factual revision.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Contriever, FiD and retrieval augmentation are inherited, not invented by Atlas. Comparative small-model/few-shot results apply to stated tasks and budgets; index updates do not guarantee accurate or complete factual revision.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• Few-shot Learning with Retrieval Augmented Language Models

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction