{"type":"link","version":"1.0","title":"Aggregating the distillation loss across M trajectory steps before each gradient update outperforms per-step optimization for long-horizon generation","author_name":"AI Archs","author_url":"https://ai-arch.pages.dev","provider_name":"AI Archs","provider_url":"https://ai-arch.pages.dev","url":"https://ai-arch.pages.dev/n/aggregated-trajectory-loss-beats-stepwise-distillation","thumbnail_url":"https://ai-arch.pages.dev/android-chrome-512x512.png","thumbnail_width":512,"thumbnail_height":512}