Qwen3-VL Models finetuned on our curated reasoning data.
🔄 In a Training Loop
Tim Ossowski
OctoMed
AI & ML interests
None yet
Organizations
None yet
Multimodal reasoning tasks for the medical domain
Popular medical benchmarks intended for zero shot evaluation (no training splits available).
Several multiple-choice text-only medical question answering tasks.
Qwen3-VL Models finetuned on our curated reasoning data.
General reasoning distilled from frontier models
Multimodal reasoning tasks for the medical domain
Medical tasks which do not have a fixed label set. Evaluation is typically done with token-f1 or other semantic similarity metrics.
Popular medical benchmarks intended for zero shot evaluation (no training splits available).
Various vision-language classification tasks for the medical domain
Several multiple-choice text-only medical question answering tasks.