Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -33,4 +33,48 @@ extra_gated_prompt: >-
|
|
33 |
extra_gated_fields:
|
34 |
By clicking Submit below I accept the terms of the license and Acceptable Use policy: checkbox
|
35 |
extra_gated_button_content: Submit
|
36 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
33 |
extra_gated_fields:
|
34 |
By clicking Submit below I accept the terms of the license and Acceptable Use policy: checkbox
|
35 |
extra_gated_button_content: Submit
|
36 |
+
---
|
37 |
+
## Links
|
38 |
+
- Website
|
39 |
+
- Paper
|
40 |
+
- Leaderboard
|
41 |
+
- Repo
|
42 |
+
|
43 |
+
## Summary
|
44 |
+
|
45 |
+
## Dataset format
|
46 |
+
The data was created by generating up to 6 obfuscations per problem for 82 problems source from original LingOly benchmark.
|
47 |
+
|
48 |
+
```python
|
49 |
+
{'question_n': # The question number in the problem
|
50 |
+
'prompt': # The main text of the question
|
51 |
+
'subprompts': [
|
52 |
+
{'questionpart_n': # For questions with several sub-parts (e.g. translating several sentences or matching paris of sentences)
|
53 |
+
'question': # The text of the question part
|
54 |
+
'answer': # The correct answer
|
55 |
+
'matches_human_scoring': # Annotation metadata
|
56 |
+
'fuzzy_matching': # Annotation metadata
|
57 |
+
'manual_edit': # Annotation metadata
|
58 |
+
}],
|
59 |
+
'metadata.preamble': # General text from the overall problem that is shared across all questions for that problem
|
60 |
+
'metadata.context': # Context text that includes important information, you should prepend your prompt with context for solvability
|
61 |
+
'metadata.obfuscated': # If this example was obfuscated or not
|
62 |
+
'metadata.obf_num': # Obfuscation counter (unique per problem). This is 0 if unobfuscated
|
63 |
+
'metadata.overall_question_n': # The problem number
|
64 |
+
'metadata.obfuscated_question_n': # Concatenation of problem number and obfuscation number
|
65 |
+
}
|
66 |
+
```
|
67 |
+
|
68 |
+
## Citation
|
69 |
+
```
|
70 |
+
@article{khouja2025lingolytoo,
|
71 |
+
title={LINGOLY-TOO: Disentangling Memorisation from Reasoning with Linguistic Templatisation and Orthographic Obfuscation},
|
72 |
+
author={Khouja, Jude and Korgul, Karolina and Hellsten, Simeon and Yang, Lingyi and Neacșu, Vlad A. and Mayne, Harry and Kearns, Ryan O. and Bean, Andrew M. and Mahdi, Adam},
|
73 |
+
year={2025},
|
74 |
+
eprint=,
|
75 |
+
archivePrefix={arXiv},
|
76 |
+
primaryClass={cs.CL},
|
77 |
+
url=,
|
78 |
+
}
|
79 |
+
|
80 |
+
```
|