ConceptZero-shot LLM Reranker

Zero-Shot LLM Reranker

A reranking method that directly uses a large language model to judge relevance or rank query-document pairs. It requires no task-specific training and offers strong generality, but incurs relatively high inference cost.

Timeline (last 90 days)

Oct 30

Compared with current state-of-the-art zero-shot LLM rerankers, LLM-IRA achieves comparable cost-effectiveness with an order of magnitude fewer LLM calls.

Unverified50%

All Facts (1)

Unverified

Compared with current state-of-the-art zero-shot LLM rerankers, LLM-IRA achieves comparable cost-effectiveness with an order of magnitude fewer LLM calls.

50%

Source Articles

LLM-IRA: Quantum-Inspired Interactive Ranking Aggregation for Boosting RAG Retrieval with Active LLM FeedbackOct 30