Background

In T375364 we created AI generated article summaries based for a list of ~8,000 popular/important articles on Wikipedia. We used a specific prompt and the Aya 32B model from Cohere. Since that time, we've updated the prompt and the model has been updated to Aya expanse. This task describes the work to re-generate those summaries using a new prompt, new model, and additional filters based on article categories.

User story

As a reader I expect AI generated summaries to be high quality and written in a tone appropriate for Wikipedia.

Requirements

New Prompt

## Task
Summarize the text below at a 7-th grade reading level, in {language}, using 100 words or less. 

## Guidelines
When writing the summary it is crucial to stick to ALL of the following guidelines:
* Simplicity: The summary should be easy to read and understand. Aim for a 7-th grade reading level; and
* Fluency: The summary should be grammatically correct. Provide only the summary itself as output and refrain from using humorous or imaginative titles; and
* Meaning preservation: The summary should be factually consistent with the input text capturing its essential points.  Do not expand content beyond its original length if it is already brief; and
* Language:  The summary should be written in {language}. If you detect yourself starting to switch to another language, stop and restart in {language}; and
* Tone: The summary should be written in a style and tone appropriate for Wikipedia. Avoid any editorializing, opinions, or expressive language about the subject and ensure the tone remains encyclopedic, neutral, and professional.

## Input text
{input_text}

New model:
Aya-expanse (tbd: hosted on either LiftWing or using the Cohere API endpoint).

Additional filters
A. Murders (crimes generally in the last 100 years)
B. Terrorist acts (e.g. hotel mumbai shooting, Las Vegas sniper)
C. Political parties that still exist
D. Terrorist groups
E. Mental health (suicide, depression)
F. Controversial subjects (bomb making; chemical weapons)

Note: the filters above have been implemented in this export script:
https://gitlab.wikimedia.org/repos/web/simple-summary-server/-/commit/7629ef97373b1dda6f7499ee70303e232c65cab7#90cc2aec681ec423b02cd67ec1df372adc43c64b_0_11

Acceptance Criteria

• New filters have been added to the summary generating criteria
• Summaries have been regenerated using new Aya expand model on Liftwing and outputted as JSON files.
• Summaries have been updated in the Extension:ArticleSummaries extension.