TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Argument Retrieval	ArguAna (BEIR)	monoT5-3B	nDCG@10	0.322	# 3
Biomedical Information Retrieval	BioASQ (BEIR)	monoT5-3B	nDCG@10	0.579	# 1
Fact Checking	CLIMATE-FEVER (BEIR)	monoT5-3B	nDCG@10	0.280	# 2
Duplicate-Question Retrieval	CQADupStack (BEIR)	monoT5-3B	nDCG@10	0.415	# 2
Entity Retrieval	DBpedia (BEIR)	monoT5-3B	nDCG@10	0.477	# 1
Fact Checking	FEVER (BEIR)	monoT5-3B	nDCG@10	0.849	# 1
Question Answering	FiQA-2018 (BEIR)	monoT5-3B	nDCG@10	0.513	# 1
Question Answering	HotpotQA (BEIR)	monoT5-3B	nDCG@10	0.759	# 1
Biomedical Information Retrieval	NFCorpus (BEIR)	monoT5-3B	nDCG@10	0.383	# 1
Question Answering	NQ (BEIR)	monoT5-3B	nDCG@10	0.633	# 2
Duplicate-Question Retrieval	Quora (BEIR)	monoT3-3B	nDCG@10	0.843	# 2
Citation Prediction	SciDocs (BEIR)	monoT5-3B	nDCG@10	0.197	# 1
Fact Checking	SciFact (BEIR)	monoT5-3B	nDCG@10	0.777	# 1
Tweet Retrieval	Signal-1M (RT) (BEIR)	monoT5-3B	nDCG@10	0.339	# 1
Argument Retrieval	Tóuche-2020 (BEIR)	monoT5-3B	nDCG@10	0.325	# 1
Biomedical Information Retrieval	TREC-COVID (BEIR)	monoT5-3B	nDCG@10	0.795	# 2
News Retrieval	TREC-NEWS (BEIR)	monoT5-3B	nDCG@10	0.473	# 2

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/biomedical-information-retrieval-on-bioasq-1)](https://paperswithcode.com/sota/biomedical-information-retrieval-on-bioasq-1?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/entity-retrieval-on-dbpedia-beir)](https://paperswithcode.com/sota/entity-retrieval-on-dbpedia-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/fact-checking-on-fever-beir)](https://paperswithcode.com/sota/fact-checking-on-fever-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/question-answering-on-fiqa-2018-beir)](https://paperswithcode.com/sota/question-answering-on-fiqa-2018-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/question-answering-on-hotpotqa-beir)](https://paperswithcode.com/sota/question-answering-on-hotpotqa-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/biomedical-information-retrieval-on-nfcorpus-1)](https://paperswithcode.com/sota/biomedical-information-retrieval-on-nfcorpus-1?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/citation-prediction-on-scidocs-beir)](https://paperswithcode.com/sota/citation-prediction-on-scidocs-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/fact-checking-on-scifact-beir)](https://paperswithcode.com/sota/fact-checking-on-scifact-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/tweet-retrieval-on-signal-1m-rt-beir)](https://paperswithcode.com/sota/tweet-retrieval-on-signal-1m-rt-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/argument-retrieval-on-touche-2020-beir)](https://paperswithcode.com/sota/argument-retrieval-on-touche-2020-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/fact-checking-on-climate-fever-beir)](https://paperswithcode.com/sota/fact-checking-on-climate-fever-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/duplicate-question-retrieval-on-cqadupstack-1)](https://paperswithcode.com/sota/duplicate-question-retrieval-on-cqadupstack-1?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/question-answering-on-nq-beir)](https://paperswithcode.com/sota/question-answering-on-nq-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/duplicate-question-retrieval-on-quora-beir)](https://paperswithcode.com/sota/duplicate-question-retrieval-on-quora-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/biomedical-information-retrieval-on-trec-1)](https://paperswithcode.com/sota/biomedical-information-retrieval-on-trec-1?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/news-retrieval-on-trec-news-beir)](https://paperswithcode.com/sota/news-retrieval-on-trec-news-beir?p=no-parameter-left-behind-how-distillation-and)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/no-parameter-left-behind-how-distillation-and/argument-retrieval-on-arguana-beir)](https://paperswithcode.com/sota/argument-retrieval-on-arguana-beir?p=no-parameter-left-behind-how-distillation-and)`

No Parameter Left Behind: How Distillation and Model Size Affect Zero-Shot Retrieval

6 Jun 2022 · Guilherme Moraes Rosa, Luiz Bonifacio, Vitor Jeronymo, Hugo Abonizio, Marzieh Fadaee, Roberto Lotufo, Rodrigo Nogueira ·

Recent work has shown that small distilled language models are strong competitors to models that are orders of magnitude larger and slower in a wide range of information retrieval tasks. This has made distilled and dense models, due to latency constraints, the go-to choice for deployment in real-world retrieval applications. In this work, we question this practice by showing that the number of parameters and early query-document interaction play a significant role in the generalization ability of retrieval models. Our experiments show that increasing model size results in marginal gains on in-domain test sets, but much larger gains in new domains never seen during fine-tuning. Furthermore, we show that rerankers largely outperform dense ones of similar size in several tasks. Our largest reranker reaches the state of the art in 12 of the 18 datasets of the Benchmark-IR (BEIR) and surpasses the previous state of the art by 3 average points. Finally, we confirm that in-domain effectiveness is not a good indicator of zero-shot effectiveness. Code is available at https://github.com/guilhermemr04/scaling-zero-shot-retrieval.git

PDF Abstract

Code

Add Remove Mark official

guilhermemr04/scaling-zero-shot-ret… official

Tasks

Add Remove

Argument Retrieval

Biomedical Information Retrieval

Citation Prediction

Duplicate-Question Retrieval

Entity Retrieval

Fact Checking

Information Retrieval

News Retrieval

Question Answering

Retrieval

Tweet Retrieval

Datasets

Natural Questions

MS MARCO

HotpotQA

DBpedia

FEVER

BEIR

BioASQ

SciDocs CLIMATE-FEVER CQADupStack FIQA NFCorpus Signal-1M

Results from the Paper

Edit

Ranked #1 on Citation Prediction on SciDocs (BEIR)

Get a GitHub badge

Task	Dataset	Model	Metric Name	Metric Value	Global Rank	Benchmark
Argument Retrieval	ArguAna (BEIR)	monoT5-3B	nDCG@10	0.322	# 3	Compare
Biomedical Information Retrieval	BioASQ (BEIR)	monoT5-3B	nDCG@10	0.579	# 1	Compare
Fact Checking	CLIMATE-FEVER (BEIR)	monoT5-3B	nDCG@10	0.280	# 2	Compare
Duplicate-Question Retrieval	CQADupStack (BEIR)	monoT5-3B	nDCG@10	0.415	# 2	Compare
Entity Retrieval	DBpedia (BEIR)	monoT5-3B	nDCG@10	0.477	# 1	Compare
Fact Checking	FEVER (BEIR)	monoT5-3B	nDCG@10	0.849	# 1	Compare
Question Answering	FiQA-2018 (BEIR)	monoT5-3B	nDCG@10	0.513	# 1	Compare
Question Answering	HotpotQA (BEIR)	monoT5-3B	nDCG@10	0.759	# 1	Compare
Biomedical Information Retrieval	NFCorpus (BEIR)	monoT5-3B	nDCG@10	0.383	# 1	Compare
Question Answering	NQ (BEIR)	monoT5-3B	nDCG@10	0.633	# 2	Compare
Duplicate-Question Retrieval	Quora (BEIR)	monoT3-3B	nDCG@10	0.843	# 2	Compare
Citation Prediction	SciDocs (BEIR)	monoT5-3B	nDCG@10	0.197	# 1	Compare
Fact Checking	SciFact (BEIR)	monoT5-3B	nDCG@10	0.777	# 1	Compare
Tweet Retrieval	Signal-1M (RT) (BEIR)	monoT5-3B	nDCG@10	0.339	# 1	Compare
Argument Retrieval	Tóuche-2020 (BEIR)	monoT5-3B	nDCG@10	0.325	# 1	Compare
Biomedical Information Retrieval	TREC-COVID (BEIR)	monoT5-3B	nDCG@10	0.795	# 2	Compare
News Retrieval	TREC-NEWS (BEIR)	monoT5-3B	nDCG@10	0.473	# 2	Compare

Methods

Add Remove

No methods listed for this paper. Add relevant methods here

Edit Social Preview

No Parameter Left Behind: How Distillation and Model Size Affect Zero-Shot Retrieval

Code Edit Add Remove Mark official

Tasks Edit Add Remove

Datasets Edit

Results from the Paper Edit

Methods Edit Add Remove

Code

Add Remove Mark official

Tasks

Add Remove

Datasets

Results from the Paper

Edit

Methods

Add Remove