A Survey of Model Architectures in Information Retrieval

By Zhichao Xu, Feng-Ran Mo, Zhi-Qi Huang, Crystina Zhang, Puxuan Yu, Bei Wang, Jimmy Lin, Vivek Srikumar. 2025

Subfields: Information Retrieval & Search, Model Architectures

Citations in the corpus

• Tevatron Meets Megatron: Expert-Parallel LLM Reranker Training on an Academic Budget (2026, unknown): “Reranking—re-scoring a first-stage candidate list with a more expensive cross-encoder (Xu et al., 2026a)—has followed the same trajectory: state-of-the-art rerankers are now billion-scale LLMs (Zhang et al., 2025), increasingly trained with knowledge distillation against strong teacher rerankers…”
• A Systematic Multi-Domain Evaluation of Document Retrievers (2026, unknown): “…meth-ods to neural dense retrievers and hybrid architectures [22, 23], surveyed dense retrieval techniques based on pre-trained language models [93], reviewed architectural developments up to modern LLM-based and generative retrieval systems [39, 88, 95], and surveyed query expansion methods [38].”
• Improving Long-Context Retrieval with Multi-Prefix Embedding (2026, unknown): “Dense retrieval encodes queries and documents as fixed-dimensional vectors for efficient inner-product search (Karpukhin et al., 2020; Izacard et al., 2022; Qwen Team, 2025; Xu et al., 2025), but compressing a long document into a single vector can obscure fine-grained evidence (Zhu et al., 2024).”