ReDiST builds student-conditioned traces for reasoning distillation
Offline selection ranks traces by answer recovery and verified-answer probability before distilling only the student model.
2 stories, newest first · one card per story
Offline selection ranks traces by answer recovery and verified-answer probability before distilling only the student model.
CoTranslator turns reverse chain-of-thought into verified forward traces, cutting distillation compute while keeping most math accuracy.
Today 1.6K readers · 2.5K views · 778 reads · all time 189K views