FEAT: Added VLSU Multimodal Dataset #1309

Open

riyosha wants to merge 8 commits intoAzure:mainfrom

riyosha:vlsu-dataset

Commits on Jan 10, 2026

Commits on Jan 12, 2026

Updated! Text and image prompts now always share the same prompt_group_id and are added/skipped together based on combined_grade. This ensures they can be sent as a single multimodal message.
riyosha
committed