[Evals] Add eval for Dhivehi diacritical marks

[aanaseer]

Thank you for contributing an eval! ♥️

🚨 Please make sure your PR follows these guidelines, failure to follow the guidelines below will result in the PR being closed automatically. Note that even if the criteria are met, that does not guarantee the PR will be merged nor GPT-4 access be granted. 🚨

PLEASE READ THIS:

In order for a PR to be merged, it must fail on GPT-4. We are aware that right now, users do not have access, so you will not be able to tell if the eval fails or not. Please run your eval with GPT-3.5-Turbo, but keep in mind as we run the eval, if GPT-4 gets higher than 90% on the eval, we will likely reject it since GPT-4 is already capable of completing the task.

We plan to roll out a way for users submitting evals to see the eval performance on GPT-4 soon. Stay tuned! Until then, you will not be able to see the eval performance on GPT-4. Starting April 10, the minimum eval count is 15 samples, we hope this makes it easier to create and contribute evals.

Also, please note that we're using Git LFS for storing the JSON files, so please make sure that you move the JSON file to Git LFS before submitting a PR. Details on how to use Git LFS are available here.

Eval details 📑

Eval name

dhivehi-diacriticals

Eval description

Given a Dhivehi sentence that includes a word missing its diacritical marks, deduce the appropriate diacritical marks for that word and output the word with correct diacritical marks.

What makes this a useful eval?

Dhivehi [1] is the Maldivian language, spoken primarily in the Maldives. It is written from right to left using Thaana [2] with vowels expressed as diacritics and consonants written on different lines. Twenty four symbols of the Thaana alphabet make up the consonants of the script [3]. There are eleven diacritics used along with the consonants. An extension of the Thaana script referred to as dotted/thikijehi Thaana [4] made up of fourteen characters is also used to transliterate Perso-Arabic script.

Consider the following sentence in Dhivehi: "މިޑްޓާމް ބަންދަށްފަހު އަލުން ސްކޫލުތަކުގެ ކިޔެވުން ފަށައިފިއެވެ" (meaning in English: Schools re-opened after midterm holidays.). In this eval, we provide the model with "މިޑްޓާމް ބަންދަށްފަހު އަލުން ސކލތކގ ކިޔެވުން ފަށައިފިއެވެ" and ask it to predict the word " ސްކޫލުތަކުގެ " with the diacritical marks.

To list a few, this eval will help evaluate the language models':
(i) robustness and adaptability to diverse linguistic systems such as those making use of right to left writing systems (specifically Dhivehi in this case)
(ii) ability to understand the phonetic and morphological properties of Dhivehi

Note: Data is from the Common Voice repo data/dv/dv_questions.txt [5]. As such, it is available under the CC0 public domain Creative Commons license [6].

References
[1] https://en.wikipedia.org/wiki/Maldivian_language
[2] https://en.wikipedia.org/wiki/Thaana
[3] https://en.wikipedia.org/wiki/Thaana#Consonants
[4] https://en.wikipedia.org/wiki/Thaana#Thikijehi_Thaana
[5] https://github.com/common-voice/common-voice/blob/main/server/data/dv/dv_questions.txt
[6] https://github.com/common-voice/common-voice/tree/main?tab=readme-ov-file#licensing-and-content-source

Criteria for a good eval ✅

Below are some of the criteria we look for in a good eval. In general, we are seeking cases where the model does not do a good job despite being capable of generating a good response (note that there are some things large language models cannot do, so those would not make good evals).

Your eval should be:

• Thematically consistent: The eval should be thematically consistent. We'd like to see a number of prompts all demonstrating some particular failure mode. For example, we can create an eval on cases where the model fails to reason about the physical world.
• Contains failures where a human can do the task, but either GPT-4 or GPT-3.5-Turbo could not.
• Includes good signal around what is the right behavior. This means either a correct answer for Basic evals or the Fact Model-graded eval, or an exhaustive rubric for evaluating answers for the Criteria Model-graded eval.
• Include at least 15 high-quality examples.

If there is anything else that makes your eval worth including, please document it below.

Unique eval value

Dhivehi is a less commonly studied language, and incorporating it into evaluations promotes linguistic diversity and inclusivity. It tests the model's capacity to learn and perform well on languages with fewer resources and less representation in the digital world, which is crucial for creating equitable AI systems.

Eval structure 🏗️

Your eval should

• Check that your data is in evals/registry/data/{name}
• Check that your YAML is registered at evals/registry/evals/{name}.yaml
• Ensure you have the right to use the data you submit via this eval

(For now, we will only be approving evals that use one of the existing eval classes. You may still write custom eval classes for your own cases, and we may consider merging them in the future.)

Final checklist 👀

Submission agreement

By contributing to Evals, you are agreeing to make your evaluation logic and data under the same MIT license as this repository. You must have adequate rights to upload any data used in an Eval. OpenAI reserves the right to use this data in future service improvements to our product. Contributions to OpenAI Evals will be subject to our usual Usage Policies (https://platform.openai.com/docs/usage-policies).

• I agree that my submission will be made available under an MIT license and complies with OpenAI's usage policies.

Email address validation

If your submission is accepted, we will be granting GPT-4 access to a limited number of contributors. Access will be given to the email address associated with the commits on the merged pull request.

• [N/A] I acknowledge that GPT-4 access will only be granted, if applicable, to the email address used for my merged pull request.

Limited availability acknowledgment

We know that you might be excited to contribute to OpenAI's mission, help improve our models, and gain access to GPT-4. However, due to the requirements mentioned above and the high volume of submissions, we will not be able to accept all submissions and thus not grant everyone who opens a PR GPT-4 access. We know this is disappointing, but we hope to set the right expectation before you open this PR.

• I understand that opening a PR, even if it meets the requirements above, does not guarantee the PR will be merged nor GPT-4 access be granted.

Submit eval

• I have filled out all required fields of this form
• I have used Git LFS for the Eval JSON data
• (Ignore if not submitting code) I have run pip install pre-commit; pre-commit install and have verified that mypy, black, isort, autoflake and ruff are running when I commit and push

Failure to fill out all required fields will result in the PR being closed.

Eval JSON data

Since we are using Git LFS, we are asking eval submitters to add in as many Eval Samples (at least 5) from their contribution here:

View evals in JSON

Eval

{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ބައެއް ތަކެތި ޗާޕުކޮށްދެއްވާނެ ފރތއ ހޯދުމަށް"}], "ideal": "ފަރާތެއް"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "އާކިން ދއކ ވާހަކައަށް އަޅާނުލުމަށް ޒޯޔާ ކައިރީ ޒެފަން ބުން"}], "ideal": "ދައްކާ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ހިނިތުންވުމުން ވީރާގެ ރތކނ އެތަށްގުނައެއް އިތުރުވެ"}], "ideal": "ރީތިކަން"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ކައިނާ ގޮސް ކައިރިއަށް ހުއްޓިލުމުންވެސް އެބިޔަ މހގ ސިފަ ވަކިވާގޮތެއްނުވި"}], "ideal": "މީހާގެ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ފައިންޕުޅެއް ނފނއ ބުނެލީވެސް ފަރުވާކުޑަ ފާޑަކަށް"}], "ideal": "ނުފެނެއޭ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "އޭ މުޙައްމަދުގެފާނެވެ! އިސްލާމާ ބެހޭ ގޮތުން ތިމަންނާއަށް ޚަބަރު ދއވ"}], "ideal": "ދެއްވާ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ލޫޝާ ހނތނވލ ނީލް ކަހަލަ ހެޔޮލަފާ ފިރިއެއް ލިބުނުކަމަށްޓަކައި ޝުކުރުވެރިވަމުން"}], "ideal": "ހިނިތުންވެލީ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "އޭރުވެސް ބަނޑަށް އަރާފައިވާ ތދށ ކުޑަކޮށްވެސް ލުޔެއް ނުލިބޭ"}], "ideal": "ތަދަށް"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ށިގަނޑުގެ ތެރޭގައި ވިންދު ޖަހަމުން ދ ހިތް އެދި ގޮވަނީ އެންމެ ކަމަކަށް"}], "ideal": "ދާ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ޔޫޝައު އެހެން ބނލމނ އައިކާ ކުޑަކޮށް ހިތް ނުތަނަވަސްވި"}], "ideal": "ބުނެލުމުން"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ކޮންމެ ގޮތަކުންވެސް އެކުވެރިޔާ ރުއްސާން ވނ"}], "ideal": "ވާނެ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ޝިފާގެ ވާހަކަ އޑއވ އަހަރެންހުރީ ބުނާނެއެއްޗެއް ނޭންގިފަ"}], "ideal": "އަޑުއިވި"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ދުވެފައި ގސ އޭނާގެ ކޮޓަރިއަށް ވަދެއްޖެ"}], "ideal": "ގޮސް"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "އެހެން ނަމަވެސް ފަސްއެނބުރި ބލކށ ޒުހުރީގެ ހިތެއް ބޭނުމެއް ނުވި"}], "ideal": "ބަލާކަށް"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "އެ ދެމީހުންގެ ސަބަބުން ފަރީނާގެ ދރއޅނވނ ނާމާންކަމާއި ބިރުވެރިކަމެއްގެ ތޫފާނަކަށް ވެއްޓިފަ"}], "ideal": "ދިރިއުޅުންވަނީ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "އަލަށް ކުރިމަތިލާ މި އމތހނގ ސަބަބުން އޭނާގެ ހިތްވަނީ ހާސްވެފަ"}], "ideal": "އިމްތިހާނުގެ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ސިޓިންގ ރޫމްގައި ތިބި އެންމެންގެ މނށ ވަކިވަކިން ބަލަމުން"}], "ideal": "މޫނަށް"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "އޭނާގެ ފޭރާމުގެ ހުދުކަން ނުހަނު ގދ"}], "ideal": "ގަދަ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "މިތަފާތު ތކގ ސަބަބުން ދިރާސާ އަށް މައުލޫމާތު އެއެްކުރަން ފަސޭހަވި"}], "ideal": "ތަކުގެ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ދެއްވާފައިވާ ސިކުނޑީގެ ބޭނުން އޭނާ ހފ ރަނގަޅު ގޮތަށް"}], "ideal": "ހިފީ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "ގޮނޑުދޮށަށް ރާޅު ބީއްސާލާ އަޑު އެ ހިސާބަށް އިވެމުން ދއއ ލުއި ލުއި މިޔުޒިކެއް ފަދައިން"}], "ideal": "ދިއައީ"}
{"input": [{"role": "system", "content": "You will be given a sentence in Dhivehi with a word missing its diacritical marks. Based on the context of the sentence, add the correct diacritical marks to the word and output only the corrected word. Remember to only output just the corrected word."}, {"role": "user", "content": "އެއީ ގާތްގަނޑަކަށް ޞަޙާބީންނަށް ފަހު އައްޑިހަ ވރކށ އަހަރު"}], "ideal": "ވަރަކަށް"}

[usama-openai]

Thanks for submitting this eval! This PR looks good. I'm approving this PR.