tokenization_robustness
Collection
26 items
•
Updated
Datasets:
Dataset Viewer
question
stringlengths 20
117
| choices
listlengths 4
4
| answer
int64 0
3
| answer_label
stringclasses 4
values | split
stringclasses 2
values | subcategories
stringclasses 5
values | lang
stringclasses 4
values | second_lang
stringclasses 2
values | coding_lang
stringclasses 1
value | notes
stringclasses 7
values | id
stringlengths 4
7
| set_id
float64 1
120
| variation_id
stringclasses 4
values |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
What type of radiation is bremsstrahlung?
|
[
"X-ray emission",
"Gamma radiation",
"electromagnetic waves",
"photon radiation"
] | 0 |
A
|
test
|
Wiki-jargon
|
eng_Latn
|
deu_Latn
|
Capitalization
|
36-0
| 36 |
0
|
|
Какую болезнь вызывает COVID-19?
|
[
"респираторное заболевание",
"болезнь сердца",
"диабет",
"рак"
] | 0 |
A
|
dev
|
Wiki-jargon
|
rus_Cyrl
|
Covid-19 russian
|
50-3.0
| 50 |
3.0
|
||
What does COVID-19 cause?
|
[
"respiratory illness",
"heart disease",
"diabetes",
"cancer"
] | 0 |
A
|
dev
|
Wiki-jargon
|
eng_Latn
|
Canonical form
|
50-1.0
| 50 |
1.0
|
||
At what conditions does Bose-Einstein condensation occur?
|
[
"bosonic interactions",
"einstein relativity",
"condensation physics",
"ultralow temperatures"
] | 3 |
D
|
dev
|
Wiki-jargon
|
eng_Latn
|
Quantum phenomenon
|
119-1.0
| 119 |
1.0
|
||
What is paracetamol?
|
[
"paracentesis",
"pain reliever",
"paracrine signaling",
"paramagnetic compound"
] | 1 |
B
|
dev
|
Wiki-jargon
|
eng_Latn
|
Drug names
|
120-1.0
| 120 |
1.0
|
||
What is acetaminophen?
|
[
"paracentesis",
"paracrine signaling",
"paramagnetic compound",
"pain reliever"
] | 3 |
D
|
dev
|
Wiki-jargon
|
eng_Latn
|
120-2.0
| 120 |
2.0
|
|||
What type of condition is Pneumonoultramicroscopicsilicovolcanoconiosis?
|
[
"eye disease",
"lung disease",
"heart disease",
"brain disease"
] | 1 |
B
|
dev
|
Rare n-grams, Wiki-jargon
|
eng_Latn
|
3-1.0
| 3 |
1.0
|
|||
นักดาราศาสตร์ชี้ให้เห็นว่าพระจันทร์เต็มดวงคืนนี้จะเหมาะสำหรับการดูดาวอย่างสมบูรณ์แบบ คืนนี้เป็นขั้นตอนใดของดวงจันทร์?
|
[
"ดวงจันทร์เต็มดวง",
"ดวงจันทร์แออัด",
"ดวงจันทร์สว่าง",
"ดวงจันทร์ยามค่ำคืน"
] | 0 |
A
|
dev
|
Cultural references, Compounds
|
tha_Thai
|
Full moon here refers to lunar phase
|
59-2.0
| 59 |
2.0
|
||
천문학자는 오늘 밤 보름달이 별 관측에 완벽하게 보일 것이라고 지적했다. 오늘 밤은 어떤 달의 위상인가?
|
[
"붐비는 달",
"완전한 달",
"밝은 달",
"밤의 달"
] | 1 |
B
|
dev
|
Cultural references, Compounds
|
kor_Hang
|
Full moon here refers to lunar phase
|
59-3.0
| 59 |
3.0
|
||
In which country is Taumatawhakatangihangakoauauotamateaturipukakapikimaungahoronukupokaiwhenuakitanatahu located?
|
[
"Australia",
"Tasmania",
"Papua New Guinea",
"New Zealand"
] | 3 |
D
|
dev
|
Rare n-grams, Regional spelling variations
|
eng_Latn
|
4-1.0
| 4 |
1.0
|
|||
What year did Eyjafjallajökull erupt and disrupt European air travel?
|
[
"2011",
"2010",
"2009",
"1982"
] | 1 |
B
|
dev
|
Rare n-grams, Regional spelling variations, Character substitution
|
eng_Latn
|
1-1.0
| 1 |
1.0
|
|||
In which country is Llanfairpwllgwyngyllgogerychwyrndrobwllllantysiliogogogoch located?
|
[
"Germany",
"France",
"Wales",
"Scotland"
] | 2 |
C
|
dev
|
Rare n-grams, Regional spelling variations, Character substitution
|
eng_Latn
|
2-1.0
| 2 |
1.0
|
Dataset Card for Tokenization Robustness
A comprehensive evaluation dataset for testing robustness of different tokenization strategies.
Dataset Details
Dataset Description
This dataset evaluates how robust language models are to different tokenization strategies and edge cases. It includes questions with multiple choice answers designed to test various aspects of tokenization handling.
- Curated by: R3
- Funded by [optional]: [More Information Needed]
- Shared by [optional]: [More Information Needed]
- Language(s) (NLP): [More Information Needed]
- License: cc
Dataset Sources [optional]
- Repository: [More Information Needed]
- Paper [optional]: [More Information Needed]
- Demo [optional]: [More Information Needed]
Uses
Direct Use
[More Information Needed]
Out-of-Scope Use
[More Information Needed]
Dataset Structure
The dataset contains multiple-choice questions with associated metadata about tokenization types and categories.
Dataset Creation
Curation Rationale
[More Information Needed]
Source Data
Data Collection and Processing
[More Information Needed]
Who are the source data producers?
[More Information Needed]
Annotations [optional]
Annotation process
[More Information Needed]
Who are the annotators?
[More Information Needed]
Personal and Sensitive Information
[More Information Needed]
Bias, Risks, and Limitations
The dataset focuses primarily on English text and may not generalize to other languages or tokenization schemes not covered in the evaluation.
Recommendations
Users should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.
Citation [optional]
BibTeX:
[More Information Needed]
APA:
[More Information Needed]
Glossary [optional]
[More Information Needed]
More Information [optional]
[More Information Needed]
Dataset Card Authors [optional]
[More Information Needed]
Dataset Card Contact
[More Information Needed]
- Downloads last month
- 10
Size of downloaded dataset files:
28.2 kB
Size of the auto-converted Parquet files:
28.2 kB
Number of rows:
32