#artificial intelligence
May 2026
LiteMedCoT-VL: Parameter-Efficient Adaptation for Medical Visual Question Answering
LiteMedCoT-VL is introduced, a pipeline that transfers chain-of-thought reasoning from a 235B teacher model to 2B student models through LoRA-based fine-tuning on explanation-enriched training data, and indicates that a 2B model with reasoning distillation can match or exceed models with twice the parameters.
Run-Ze Ma, Shun-Bo Jia, Haonan Lyu et al.
· arXiv.org · 0 citations