Open-Domain Question Answering

Open-domain question answering is the task of question answering on open-domain datasets such as Wikipedia.

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 301–325 of 494 papers

Title	Date	Tasks	Status
Targeted Extraction of Temporal Facts from Textual Resources for Improved Temporal Question Answering over Knowledge Bases	Mar 21, 2022	Knowledge Base Question AnsweringOpen-Domain Question Answering	—Unverified
Calibration of Machine Reading Systems at Scale	Mar 20, 2022	Claim VerificationOpen-Domain Question Answering	—Unverified
C-MORE: Pretraining to Answer Open-Domain Questions by Consulting Millions of References	Mar 16, 2022	Open-Domain Question AnsweringQuestion Answering	CodeCode Available
Internet-augmented language models through few-shot prompting for open-domain question answering	Mar 10, 2022	Language ModellingOpen-Domain Question Answering	—Unverified
Semantic Sentence Composition Reasoning for Multi-Hop Question Answering	Mar 1, 2022	Multi-hop Question AnsweringOpen-Domain Question Answering	—Unverified
Can Open Domain Question Answering Systems Answer Visual Knowledge Questions?	Feb 9, 2022	Open-Domain Question AnsweringQuestion Answering	—Unverified
Improving Biomedical Information Retrieval with Neural Retrievers	Jan 19, 2022	Biomedical Information RetrievalInformation Retrieval	—Unverified
Evidentiality-guided Generation for Knowledge-Intensive NLP Tasks	Jan 16, 2022	Fact VerificationMemorization	—Unverified
In Situ Answer Sentence Selection at Web-scale	Jan 16, 2022	Multi-Task LearningOpen-Domain Question Answering	—Unverified
An Encoder Attribution Analysis for Dense Passage Retriever in Open-Domain Question Answering	Jan 16, 2022	Open-Domain Question AnsweringQuestion Answering	—Unverified
Reasoning over Hybrid Chain for Table-and-Text Open Domain QA	Jan 15, 2022	Open-Domain Question AnsweringQuestion Answering	—Unverified
Tackling Query-Focused Summarization as A Knowledge-Intensive Task: A Pilot Study	Dec 14, 2021	Open-Domain Question AnsweringPassage Retrieval	—Unverified
You Only Need One Model for Open-domain Question Answering	Dec 14, 2021	Hard AttentionNatural Questions	—Unverified
Quality and Cost Trade-offs in Passage Re-ranking Task	Nov 18, 2021	Information RetrievalOpen-Domain Question Answering	—Unverified
Get Your Model Puzzled: Introducing Crossword-Solving as a New NLP Benchmark	Nov 16, 2021	Natural Language UnderstandingOpen-Domain Question Answering	—Unverified
Unsupervised Open-Domain Question Answering with Higher Answerability	Nov 16, 2021	Machine Reading ComprehensionOpen-Domain Question Answering	—Unverified
CCQA: A New Web-Scale Question Answering Dataset for Model Pre-Training	Nov 16, 2021	Open-Domain Question AnsweringQuestion Answering	—Unverified
Calibration of Machine Reading Systems at Scale	Nov 16, 2021	Claim VerificationOpen-Domain Question Answering	—Unverified
Reasoning over Hybrid Chain for Table-and-Text Open Domain Question Answering	Nov 16, 2021	Open-Domain Question AnsweringQuestion Answering	—Unverified
Salient Phrase Aware Dense Retrieval: Can a Dense Retriever Imitate a Sparse One?	Nov 16, 2021	Open-Domain Question AnsweringPassage Retrieval	—Unverified
Open Domain Question Answering over Virtual Documents: A Unified Approach for Data and Text	Nov 16, 2021	Data-to-Text GenerationNatural Questions	—Unverified
ArchivalQA: A Large-scale Benchmark Dataset for Open Domain Question Answering over Archival News Collections	Nov 16, 2021	Open-Domain Question AnsweringQuestion Answering	—Unverified
Hyperlink-induced Pre-training for Passage Retrieval of Open-domain Question Answering	Nov 16, 2021	Open-Domain Question AnsweringPassage Retrieval	—Unverified
Simple and Effective Unsupervised Redundancy Elimination to Compress Dense Vectors for Passage Retrieval	Nov 1, 2021	Open-Domain Question AnsweringPassage Retrieval	—Unverified
Multi-Task Dense Retrieval via Model Uncertainty Fusion for Open-Domain Question Answering	Nov 1, 2021	ArticlesOpen-Domain Question Answering	CodeCode Available

Show:10 25 50

← PrevPage 13 of 20Next →

All datasets KILT: ELI5 KILT: Natural Questions KILT: TriviaQA KILT: HotpotQA SearchQA ELI5 QUASAR Natural Questions SQuAD1.1 dev WebQuestions SQuAD1.1 DuReader

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	somebody	KILT-RL	2.62	—	Unverified
2	Wikipedia	KILT-RL	2.46	—	Unverified
3	arxiv.org/abs/2103.06332	KILT-RL	2.36	—	Unverified
4	BART + DPR	KILT-RL	1.9	—	Unverified
5	RAG	KILT-RL	1.69	—	Unverified
6	Training Set Retrieval (top 1)	KILT-RL	0	—	Unverified
7	T5-base	KILT-RL	0	—	Unverified
8	Input Copying	KILT-RL	0	—	Unverified
9	Sphere	KILT-RL	0	—	Unverified
10	Random Training Set Answer	KILT-RL	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Re2G	KILT-EM	43.56	—	Unverified
2	intersect	KILT-EM	38.78	—	Unverified
3	KGI_0	KILT-EM	36.36	—	Unverified
4	Wikipedia	KILT-EM	35.32	—	Unverified
5	RAG	KILT-EM	32.69	—	Unverified
6	BERT + DPR	KILT-EM	31.99	—	Unverified
7	BART + DPR	KILT-EM	30.06	—	Unverified
8	Multitask DPR + BART	KILT-EM	29.09	—	Unverified
9	Multi-task DPR	KILT-EM	0	—	Unverified
10	Sphere	KILT-EM	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Re2G	KILT-EM	57.91	—	Unverified
2	intersect	KILT-EM	50.56	—	Unverified
3	Wikipedia	KILT-EM	45.55	—	Unverified
4	KGI_0	KILT-EM	42.85	—	Unverified
5	Multitask DPR + BART	KILT-EM	42.36	—	Unverified
6	RAG	KILT-EM	38.13	—	Unverified
7	BERT + DPR	KILT-EM	34.48	—	Unverified
8	BART + DPR	KILT-EM	31.4	—	Unverified
9	TABi	KILT-EM	0	—	Unverified
10	T5-base	KILT-EM	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	intersect	KILT-EM	18.06	—	Unverified
2	Wikipedia	KILT-EM	11.71	—	Unverified
3	Multitask DPR + BART	KILT-EM	9.53	—	Unverified
4	RAG	KILT-EM	3.21	—	Unverified
5	BART + DPR	KILT-EM	1.96	—	Unverified
6	BERT + DPR	KILT-EM	0.74	—	Unverified
7	Sphere	KILT-EM	0	—	Unverified
8	Multi-task DPR	KILT-EM	0	—	Unverified
9	GENRE	KILT-EM	0	—	Unverified
10	chriskuei	KILT-EM	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SpanBERT	F1	84.8	—	Unverified
2	Cluster-Former (#C=512)	EM	68	—	Unverified
3	Locality-Sensitive Hashing	EM	66	—	Unverified
4	Multi-passage BERT	EM	65.1	—	Unverified
5	Sparse Attention	EM	64.7	—	Unverified
6	DECAPROP	EM	62.2	—	Unverified
7	Bi-Attention + DCU-LSTM	N-gram F1	59.5	—	Unverified
8	Denoising QA	EM	58.8	—	Unverified
9	DecaProp	EM	56.8	—	Unverified
10	AMANDA	N-gram F1	56.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Fourier Transformer	Rouge-L	26.9	—	Unverified
2	QG	Rouge-L	26.4	—	Unverified
3	BART	Rouge-L	24.3	—	Unverified
4	E-MCA	Rouge-L	24	—	Unverified
5	Transformer Multitask + LayerDrop	Rouge-L	23.4	—	Unverified
6	Multi-Inrerleave	Rouge-L	14.63	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Evidence Aggregation via R^3 Re-Ranking	EM (Quasar-T)	42.3	—	Unverified
2	Denoising QA	EM (Quasar-T)	42.2	—	Unverified
3	DecaProp	EM (Quasar-T)	38.6	—	Unverified
4	R^3	EM (Quasar-T)	35.3	—	Unverified
5	GA	EM (Quasar-T)	26.4	—	Unverified
6	BiDAF	EM (Quasar-T)	25.9	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	FiE	Exact Match	58.4	—	Unverified
2	R2-D2 HN-DPR	Exact Match	55.9	—	Unverified
3	UniK-QA	Exact Match	54.9	—	Unverified
4	UnitedQA (Hybrid)	Exact Match	54.7	—	Unverified
5	BPR (linear scan; l=1000)	Exact Match	41.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SPARTA	EM	59.3	—	Unverified
2	Blended RAG	EM	57.63	—	Unverified
3	BERTserini	EM	50.2	—	Unverified
4	BERTserini	EM	38.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	UniK-QA	Exact Match	57.7	—	Unverified
2	FiE+PAQ	Exact Match	56.3	—	Unverified
3	FiE	Exact Match	52.4	—	Unverified
4	EMDR2	Exact Match	48.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DrQA	EM	70	—	Unverified
2	DCN	EM	66.2	—	Unverified
3	MPCM	EM	65.5	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	ERNIE 2.0 Large	EM	64.2	—	Unverified
2	ERNIE 2.0 Base	EM	61.3	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	UniK-QA	Exact Match	65.5	—	Unverified
2	BPR (linear scan; l=1000)	Exact Match	56.8	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	EMDR2	Exact Match	52.5	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	UnitedQA (Hybrid)	Exact Match	70.5	—	Unverified