AI RESEARCH
Fluent Alignment with Disfluent Judges: Post-training for Lower-resource Languages
arXiv CS.AI
•
ArXi:2512.08777v2 Announce Type: replace-cross We propose a post-