longvila-scaling-rl-long-videos

【效率与协作】Train vision-language models on hour-long videos using a two-stage pipeline combining supervised fine-tuning with GRPO, reaching 71% accuracy on VideoMME while supporting 8,192 frames through efficient multi-modal sequence parallelism.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准