LLM2IR: simple unsupervised contrastive learning makes long-context LLM great retriever

Xiaocong Yang

LLM2IR: simple unsupervised contrastive learning makes long-context LLM great retriever

Information Retrieval 2026-01-12 v1 Artificial Intelligence Computation and Language

Authors: Xiaocong Yang

Abstract

Modern dense information retrieval (IR) models usually rely on costly large-scale pretraining. In this paper, we introduce LLM2IR, an efficient unsupervised contrastive learning framework to convert any decoder-only large language model (LLM) to an information retrieval model. Despite its simplicity, the effectiveness is proven among different LLMs on multiple IR benchmarks including LoCo, LongEmbed and BEIR. We also find that models with a longer context length tend to have a stronger IR capacity by comparing task performances of models in the same model family. Our work not only provides an effective way to build IR models on the state-of-the-art LLMs, but also shed light on the relationship between information retrieval ability and model context length, which helps the design of better information retrievers.

Keywords

long-context modeling in-context learning contrastive learning

Cite

@article{arxiv.2601.05262,
  title  = {LLM2IR: simple unsupervised contrastive learning makes long-context LLM great retriever},
  author = {Xiaocong Yang},
  journal= {arXiv preprint arXiv:2601.05262},
  year   = {2026}
}

Comments

MS Thesis

LLM2IR: simple unsupervised contrastive learning makes long-context LLM great retriever

Abstract

Keywords

Cite

Comments

Related papers