Syntriever: How to Train Your Retriever with Synthetic Data from LLMs

By Minsang Kim, Seung Baek. 2025

Subfields: Information Retrieval & Search, Large Language Models

Citations in the corpus

• ExpandR: Teaching Dense Retrievers Beyond Queries with LLM Guidance (2025, unknown): “The first direction leverages LLM-generated reformulations as supervision signals to train dense retrieval models, typically through contrastive training (Zhang et al., 2025; Ma et al., 2025) or ranking probability distillation (Shi et al., 2024; Kim and Baek, 2025).”