PyTerrier-GenRank: The PyTerrier Plugin for Reranking with Large Language Models

Kaustubh D. Dhole

PyTerrier-GenRank: The PyTerrier Plugin for Reranking with Large Language Models

Information Retrieval 2024-12-10 v1 Artificial Intelligence Computation and Language

Authors: Kaustubh D. Dhole

Abstract

Using LLMs as rerankers requires experimenting with various hyperparameters, such as prompt formats, model choice, and reformulation strategies. We introduce PyTerrier-GenRank, a PyTerrier plugin to facilitate seamless reranking experiments with LLMs, supporting popular ranking strategies like pointwise and listwise prompting. We validate our plugin through HuggingFace and OpenAI hosted endpoints.

Keywords

large language model information retrieval

Cite

@article{arxiv.2412.05339,
  title  = {PyTerrier-GenRank: The PyTerrier Plugin for Reranking with Large Language Models},
  author = {Kaustubh D. Dhole},
  journal= {arXiv preprint arXiv:2412.05339},
  year   = {2024}
}

Related papers

View all related →