Published on August 17, 2024, the article presents a retrieval method that iteratively generates queries for multi-hop questions without requiring a language-model call at every iteration.
The article on EfficientRAG describes a retriever for multi-hop questions: it generates new queries iteratively without requiring a language-model call at every step. The proposal aims to reduce repeated calls for complex questions; the supplied material provides no quantitative results or further evaluation details.
To consult and verify the source, search for the article titled “EfficientRAG para perguntas multi-hop” and check its publication date, August 17, 2024. Review the original for how the method generates queries and what experiments, metrics, and limitations it reports. If you use AI to study or apply the idea, do not submit personal data or confidential documents without appropriate authorization and policy.