Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

RankVicuna: Zero-Shot Listwise Document Reranking with Open-Source Large Language Models

About

Researchers have successfully applied large language models (LLMs) such as ChatGPT to reranking in an information retrieval context, but to date, such work has mostly been built on proprietary models hidden behind opaque API endpoints. This approach yields experimental results that are not reproducible and non-deterministic, threatening the veracity of outcomes that build on such shaky foundations. To address this significant shortcoming, we present RankVicuna, the first fully open-source LLM capable of performing high-quality listwise reranking in a zero-shot setting. Experimental results on the TREC 2019 and 2020 Deep Learning Tracks show that we can achieve effectiveness comparable to zero-shot reranking with GPT-3.5 with a much smaller 7B parameter model, although our effectiveness remains slightly behind reranking with GPT-4. We hope our work provides the foundation for future research on reranking with modern LLMs. All the code necessary to reproduce our results is available at https://github.com/castorini/rank_llm.

Ronak Pradeep, Sahel Sharifymoghaddam, Jimmy Lin• 2023

Related benchmarks

TaskDatasetResultRank
Document RankingTREC DL Track 2019 (test)
nDCG@1068.9
96
RerankingTREC 2020 (test)
NDCG@1066.1
55
Question AnsweringScientific QA Base setting
F1 Score45.27
38
Information RetrievalScientific QA Base setting
HitRate@152.53
38
RankingBEIR selected subset v1.0.0 (test)
TREC-COVID80.5
38
End-to-end Question Answering2WikiMultiHopQA (test val)
EM32.66
20
End-to-end Question AnsweringHotpotQA (test val)
EM32.08
20
End-to-end Question AnsweringMuSiQue (test val)
EM7.78
20
End-to-end Question AnsweringMultiHopRAG (test val)
Accuracy42.76
20
RerankingSciRAG-SSLI hard 1.0 (test)
Hit Rate @ 158.67
19
Showing 10 of 15 rows

Other info

Follow for update