wujoe132/ponys-ai-multilingual-companion-evaluation
Ponys.ai Multilingual AI Companion Evaluation Protocols This public collection contains 20 reusable evaluation protocols for AI companion and character experiences. It covers conversation memory, persona consistency, consent recovery, visual continuity, code switching, and regional language behavior across Japanese, Korean, Latin American Spanish, Brazilian Portuguese, Simplified Chinese, Traditional Chinese, and English. Each protocol includes structured metadata and a CSV… See the full description on the dataset page: https://huggingface.co/datasets/wujoe132/ponys-ai-multilingual-companion-evaluation.
Ponys.ai Multilingual AI Companion Evaluation Protocols
This public collection contains 20 reusable evaluation protocols for AI companion and character experiences. It covers conversation memory, persona consistency, consent recovery, visual continuity, code switching, and regional language behavior across Japanese, Korean, Latin American Spanish, Brazilian Portuguese, Simplified Chinese, Traditional Chinese, and English.
Each protocol includes structured metadata and a CSV worksheet. These are blank testing instruments for research, reviews, and product QA, not product performance claims.
Official resources
Regional coverage
- Japan: keigo boundary repair, persona drift, visual continuity, and consent reset.
- Korea: honorific consistency, roleplay memory, image identity drift, and private-mode expectations.
- Latin America: slang retention, conversational consent, long-session character memory, and image-prompt fidelity.
- Brazil: register shift, roleplay memory, boundary recovery, and character image consistency.
- Chinese and global: tone memory, relationship state, dialogue coherence, and multilingual code-switch recovery.
Citation and license
Cite as Ponys.ai Multilingual AI Companion Evaluation Protocols, version 1.0.0. Licensed under CC BY 4.0.