nthakur commited on
Commit
c64bcaa
1 Parent(s): bcfb1db

Initial add

Browse files
Files changed (1) hide show
  1. README.md +285 -0
README.md ADDED
@@ -0,0 +1,285 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ annotations_creators: []
3
+ language_creators: []
4
+ languages:
5
+ - en
6
+ licenses:
7
+ - cc-by-sa-4.0
8
+ multilinguality:
9
+ - monolingual
10
+ paperswithcode_id: beir
11
+ pretty_name: BEIR Benchmark
12
+ size_categories:
13
+ msmarco:
14
+ - 1M<n<10M
15
+ trec-covid:
16
+ - 100k<n<1M
17
+ nfcorpus:
18
+ - 1K<n<10K
19
+ nq:
20
+ - 1M<n<10M
21
+ hotpotqa:
22
+ - 1M<n<10M
23
+ fiqa:
24
+ - 10K<n<100K
25
+ arguana:
26
+ - 1K<n<10K
27
+ touche-2020:
28
+ - 100K<n<1M
29
+ cqadupstack:
30
+ - 100K<n<1M
31
+ quora:
32
+ - 100K<n<1M
33
+ dbpedia:
34
+ - 1M<n<10M
35
+ scidocs:
36
+ - 10K<n<100K
37
+ fever:
38
+ - 1M<n<10M
39
+ climate-fever:
40
+ - 1M<n<10M
41
+ scifact:
42
+ - 1K<n<10K
43
+ source_datasets: []
44
+ task_categories:
45
+ - text-retrieval
46
+ - zero-shot-retrieval
47
+ - information-retrieval
48
+ - zero-shot-information-retrieval
49
+ task_ids:
50
+ - passage-retrieval
51
+ - entity-linking-retrieval
52
+ - fact-checking-retrieval
53
+ - tweet-retrieval
54
+ - citation-prediction-retrieval
55
+ - duplication-question-retrieval
56
+ - argument-retrieval
57
+ - news-retrieval
58
+ - biomedical-information-retrieval
59
+ - question-answering-retrieval
60
+ ---
61
+
62
+ # Dataset Card for BEIR Benchmark
63
+
64
+ ## Table of Contents
65
+ - [Dataset Description](#dataset-description)
66
+ - [Dataset Summary](#dataset-summary)
67
+ - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
68
+ - [Languages](#languages)
69
+ - [Dataset Structure](#dataset-structure)
70
+ - [Data Instances](#data-instances)
71
+ - [Data Fields](#data-fields)
72
+ - [Data Splits](#data-splits)
73
+ - [Dataset Creation](#dataset-creation)
74
+ - [Curation Rationale](#curation-rationale)
75
+ - [Source Data](#source-data)
76
+ - [Annotations](#annotations)
77
+ - [Personal and Sensitive Information](#personal-and-sensitive-information)
78
+ - [Considerations for Using the Data](#considerations-for-using-the-data)
79
+ - [Social Impact of Dataset](#social-impact-of-dataset)
80
+ - [Discussion of Biases](#discussion-of-biases)
81
+ - [Other Known Limitations](#other-known-limitations)
82
+ - [Additional Information](#additional-information)
83
+ - [Dataset Curators](#dataset-curators)
84
+ - [Licensing Information](#licensing-information)
85
+ - [Citation Information](#citation-information)
86
+ - [Contributions](#contributions)
87
+
88
+ ## Dataset Description
89
+
90
+ - **Homepage:** https://github.com/UKPLab/beir
91
+ - **Repository:** https://github.com/UKPLab/beir
92
+ - **Paper:** https://openreview.net/forum?id=wCu6T5xFjeJ
93
+ - **Leaderboard:** https://docs.google.com/spreadsheets/d/1L8aACyPaXrL8iEelJLGqlMqXKPX2oSP_R10pZoy77Ns
94
+ - **Point of Contact:** nandan.thakur@uwaterloo.ca
95
+
96
+ ### Dataset Summary
97
+
98
+ BEIR is a heterogeneous benchmark that has been built from 18 diverse datasets representing 9 information retrieval tasks:
99
+
100
+ - Fact-checking: [FEVER](http://fever.ai), [Climate-FEVER](http://climatefever.ai), [SciFact](https://github.com/allenai/scifact)
101
+ - Question-Answering: [NQ](https://ai.google.com/research/NaturalQuestions), [HotpotQA](https://hotpotqa.github.io), [FiQA-2018](https://sites.google.com/view/fiqa/)
102
+ - Bio-Medical IR: [TREC-COVID](https://ir.nist.gov/covidSubmit/index.html), [BioASQ](http://bioasq.org), [NFCorpus](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/)
103
+ - News Retrieval: [TREC-NEWS](https://trec.nist.gov/data/news2019.html), [Robust04](https://trec.nist.gov/data/robust/04.guidelines.html)
104
+ - Argument Retrieval: [Touche-2020](https://webis.de/events/touche-20/shared-task-1.html), [ArguAna](tp://argumentation.bplaced.net/arguana/data)
105
+ - Duplicate Question Retrieval: [Quora](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs), [CqaDupstack](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/)
106
+ - Citation-Prediction: [SCIDOCS](https://allenai.org/data/scidocs)
107
+ - Tweet Retrieval: [Signal-1M](https://research.signal-ai.com/datasets/signal1m-tweetir.html)
108
+ - Entity Retrieval: [DBPedia](https://github.com/iai-group/DBpedia-Entity/)
109
+
110
+ All these datasets have been preprocessed and can be used for your experiments.
111
+
112
+
113
+ ```python
114
+
115
+ ```
116
+
117
+ ### Supported Tasks and Leaderboards
118
+
119
+ The dataset supports a leaderboard that evaluates models against task-specific metrics such as F1 or EM, as well as their ability to retrieve supporting information from Wikipedia.
120
+
121
+ The current best performing models can be found [here](https://eval.ai/web/challenges/challenge-page/689/leaderboard/).
122
+
123
+ ### Languages
124
+
125
+ All tasks are in English (`en`).
126
+
127
+ ## Dataset Structure
128
+
129
+ All BEIR datasets must contain a corpus, queries and qrels (relevance judgments file). They must be in the following format:
130
+ - `corpus` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with three fields `_id` with unique document identifier, `title` with document title (optional) and `text` with document paragraph or passage. For example: `{"_id": "doc1", "title": "Albert Einstein", "text": "Albert Einstein was a German-born...."}`
131
+ - `queries` file: a `.jsonl` file (jsonlines) that contains a list of dictionaries, each with two fields `_id` with unique query identifier and `text` with query text. For example: `{"_id": "q1", "text": "Who developed the mass-energy equivalence formula?"}`
132
+ - `qrels` file: a `.tsv` file (tab-seperated) that contains three columns, i.e. the `query-id`, `corpus-id` and `score` in this order. Keep 1st row as header. For example: `q1 doc1 1`
133
+
134
+ ### Data Instances
135
+
136
+ A high level example of any beir dataset:
137
+
138
+ ```python
139
+ corpus = {
140
+ "doc1" : {
141
+ "title": "Albert Einstein",
142
+ "text": "Albert Einstein was a German-born theoretical physicist. who developed the theory of relativity, \
143
+ one of the two pillars of modern physics (alongside quantum mechanics). His work is also known for \
144
+ its influence on the philosophy of science. He is best known to the general public for his mass–energy \
145
+ equivalence formula E = mc2, which has been dubbed 'the world's most famous equation'. He received the 1921 \
146
+ Nobel Prize in Physics 'for his services to theoretical physics, and especially for his discovery of the law \
147
+ of the photoelectric effect', a pivotal step in the development of quantum theory."
148
+ },
149
+ "doc2" : {
150
+ "title": "", # Keep title an empty string if not present
151
+ "text": "Wheat beer is a top-fermented beer which is brewed with a large proportion of wheat relative to the amount of \
152
+ malted barley. The two main varieties are German Weißbier and Belgian witbier; other types include Lambic (made\
153
+ with wild yeast), Berliner Weisse (a cloudy, sour beer), and Gose (a sour, salty beer)."
154
+ },
155
+ }
156
+
157
+ queries = {
158
+ "q1" : "Who developed the mass-energy equivalence formula?",
159
+ "q2" : "Which beer is brewed with a large proportion of wheat?"
160
+ }
161
+
162
+ qrels = {
163
+ "q1" : {"doc1": 1},
164
+ "q2" : {"doc2": 1},
165
+ }
166
+ ```
167
+
168
+ ### Data Fields
169
+
170
+ Examples from all configurations have the following features:
171
+
172
+ ### Corpus
173
+ - `corpus`: a `dict` feature representing the document title and passage text, made up of:
174
+ - `_id`: a `string` feature representing the unique document id
175
+ - `title`: a `string` feature, denoting the title of the document.
176
+ - `text`: a `string` feature, denoting the text of the document.
177
+
178
+ ### Queries
179
+ - `queries`: a `dict` feature representing the query, made up of:
180
+ - `_id`: a `string` feature representing the unique query id
181
+ - `text`: a `string` feature, denoting the text of the query.
182
+
183
+ ### Qrels
184
+ - `qrels`: a `dict` feature representing the query document relevance judgements, made up of:
185
+ - `_id`: a `string` feature representing the query id
186
+ - `_id`: a `string` feature, denoting the document id.
187
+ - `score`: a `int32` feature, denoting the relevance judgement between query and document.
188
+
189
+
190
+ ### Data Splits
191
+
192
+ | Dataset | Website| BEIR-Name | Type | Queries | Corpus | Rel D/Q | Down-load | md5 |
193
+ | -------- | -----| ---------| --------- | ----------- | ---------| ---------| :----------: | :------:|
194
+ | MSMARCO | [Homepage](https://microsoft.github.io/msmarco/)| ``msmarco`` | ``train``<br>``dev``<br>``test``| 6,980 | 8.84M | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/msmarco.zip) | ``444067daf65d982533ea17ebd59501e4`` |
195
+ | TREC-COVID | [Homepage](https://ir.nist.gov/covidSubmit/index.html)| ``trec-covid``| ``test``| 50| 171K| 493.5 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/trec-covid.zip) | ``ce62140cb23feb9becf6270d0d1fe6d1`` |
196
+ | NFCorpus | [Homepage](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/) | ``nfcorpus`` | ``train``<br>``dev``<br>``test``| 323 | 3.6K | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nfcorpus.zip) | ``a89dba18a62ef92f7d323ec890a0d38d`` |
197
+ | BioASQ | [Homepage](http://bioasq.org) | ``bioasq``| ``train``<br>``test`` | 500 | 14.91M | 8.05 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#2-bioasq) |
198
+ | NQ | [Homepage](https://ai.google.com/research/NaturalQuestions) | ``nq``| ``train``<br>``test``| 3,452 | 2.68M | 1.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/nq.zip) | ``d4d3d2e48787a744b6f6e691ff534307`` |
199
+ | HotpotQA | [Homepage](https://hotpotqa.github.io) | ``hotpotqa``| ``train``<br>``dev``<br>``test``| 7,405 | 5.23M | 2.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/hotpotqa.zip) | ``f412724f78b0d91183a0e86805e16114`` |
200
+ | FiQA-2018 | [Homepage](https://sites.google.com/view/fiqa/) | ``fiqa`` | ``train``<br>``dev``<br>``test``| 648 | 57K | 2.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fiqa.zip) | ``17918ed23cd04fb15047f73e6c3bd9d9`` |
201
+ | Signal-1M(RT) | [Homepage](https://research.signal-ai.com/datasets/signal1m-tweetir.html)| ``signal1m`` | ``test``| 97 | 2.86M | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#4-signal-1m) |
202
+ | TREC-NEWS | [Homepage](https://trec.nist.gov/data/news2019.html) | ``trec-news`` | ``test``| 57 | 595K | 19.6 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#1-trec-news) |
203
+ | ArguAna | [Homepage](http://argumentation.bplaced.net/arguana/data) | ``arguana``| ``test`` | 1,406 | 8.67K | 1.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/arguana.zip) | ``8ad3e3c2a5867cdced806d6503f29b99`` |
204
+ | Touche-2020| [Homepage](https://webis.de/events/touche-20/shared-task-1.html) | ``webis-touche2020``| ``test``| 49 | 382K | 19.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/webis-touche2020.zip) | ``46f650ba5a527fc69e0a6521c5a23563`` |
205
+ | CQADupstack| [Homepage](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/) | ``cqadupstack``| ``test``| 13,145 | 457K | 1.4 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/cqadupstack.zip) | ``4e41456d7df8ee7760a7f866133bda78`` |
206
+ | Quora| [Homepage](https://www.quora.com/q/quoradata/First-Quora-Dataset-Release-Question-Pairs) | ``quora``| ``dev``<br>``test``| 10,000 | 523K | 1.6 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/quora.zip) | ``18fb154900ba42a600f84b839c173167`` |
207
+ | DBPedia | [Homepage](https://github.com/iai-group/DBpedia-Entity/) | ``dbpedia-entity``| ``dev``<br>``test``| 400 | 4.63M | 38.2 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/dbpedia-entity.zip) | ``c2a39eb420a3164af735795df012ac2c`` |
208
+ | SCIDOCS| [Homepage](https://allenai.org/data/scidocs) | ``scidocs``| ``test``| 1,000 | 25K | 4.9 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scidocs.zip) | ``38121350fc3a4d2f48850f6aff52e4a9`` |
209
+ | FEVER | [Homepage](http://fever.ai) | ``fever``| ``train``<br>``dev``<br>``test``| 6,666 | 5.42M | 1.2| [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/fever.zip) | ``5a818580227bfb4b35bb6fa46d9b6c03`` |
210
+ | Climate-FEVER| [Homepage](http://climatefever.ai) | ``climate-fever``|``test``| 1,535 | 5.42M | 3.0 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/climate-fever.zip) | ``8b66f0a9126c521bae2bde127b4dc99d`` |
211
+ | SciFact| [Homepage](https://github.com/allenai/scifact) | ``scifact``| ``train``<br>``test``| 300 | 5K | 1.1 | [Link](https://public.ukp.informatik.tu-darmstadt.de/thakur/BEIR/datasets/scifact.zip) | ``5f7d1de60b170fc8027bb7898e2efca1`` |
212
+ | Robust04 | [Homepage](https://trec.nist.gov/data/robust/04.guidelines.html) | ``robust04``| ``test``| 249 | 528K | 69.9 | No | [How to Reproduce?](https://github.com/UKPLab/beir/blob/main/examples/dataset#3-robust04) |
213
+
214
+
215
+ ## Dataset Creation
216
+
217
+ ### Curation Rationale
218
+
219
+ [Needs More Information]
220
+
221
+ ### Source Data
222
+
223
+ #### Initial Data Collection and Normalization
224
+
225
+ [Needs More Information]
226
+
227
+ #### Who are the source language producers?
228
+
229
+ [Needs More Information]
230
+
231
+ ### Annotations
232
+
233
+ #### Annotation process
234
+
235
+ [Needs More Information]
236
+
237
+ #### Who are the annotators?
238
+
239
+ [Needs More Information]
240
+
241
+ ### Personal and Sensitive Information
242
+
243
+ [Needs More Information]
244
+
245
+ ## Considerations for Using the Data
246
+
247
+ ### Social Impact of Dataset
248
+
249
+ [Needs More Information]
250
+
251
+ ### Discussion of Biases
252
+
253
+ [Needs More Information]
254
+
255
+ ### Other Known Limitations
256
+
257
+ [Needs More Information]
258
+
259
+ ## Additional Information
260
+
261
+ ### Dataset Curators
262
+
263
+ [Needs More Information]
264
+
265
+ ### Licensing Information
266
+
267
+ [Needs More Information]
268
+
269
+ ### Citation Information
270
+
271
+ Cite as:
272
+ ```
273
+ @inproceedings{
274
+ thakur2021beir,
275
+ title={{BEIR}: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models},
276
+ author={Nandan Thakur and Nils Reimers and Andreas R{\"u}ckl{\'e} and Abhishek Srivastava and Iryna Gurevych},
277
+ booktitle={Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2)},
278
+ year={2021},
279
+ url={https://openreview.net/forum?id=wCu6T5xFjeJ}
280
+ }
281
+ ```
282
+
283
+ ### Contributions
284
+
285
+ Thanks to [@Nthakur20](https://github.com/Nthakur20) for adding this dataset.