Skip to content
Open source

How researchers adapted Dolma for better Thai language models.

Ai2 BlogResearch source·
AI brief

Thai researchers adapted Ai2's open Dolma toolkit to build Mangosteen, a 47-billion-token Thai corpus that filters low-quality web data while maintaining or improving model performance.

Why it matters: The corpus maintains or improves model performance while strengthening Thai cultural knowledge.

Written by AI from Ai2 Blog's published text.

Source

“Thai researchers adapted Ai2’s open Dolma toolkit to build Mangosteen, a 47-billion-token Thai corpus that filters low-quality web data while maintaining or improving model performance and strengthening Thai cultural knowledge.”
Ai2 Blog, 26 Aug 2026, 08:00.

A paper, laboratory publication or research report.

Read original story ↗