SOURCE-LINKED INTELLIGENCE

ReDraft, Don't Just Distill: Reference-Driven Revision for Continual VLLM Post-Training

arXiv · AI, language, vision and robotics · article · Sep 15, 2026 · UTC

Continual post-training of large multimodal models should add new capabilities while preserving those from pre-training, and the two goals pull in opposite directions. SFT gives explicit target supervision that learns a task from near-zero accuracy, but its off-policy targets move the model far enough to cause forgetting; on-policy methods such as RLVR and self-distillation preserve policy proximity yet supply little signal when the policy cannot yet solve the task. We introduce ReDraft (Reference-Driven Revision and Fine-Tuning), which obtains both from the model's own failures: using an expe

Read original source ↗ Open in workspace

recordType

paper

region

Global