ji4chenli / t2v-turbo-v2

Enhancing Video Model Post-Training through Data, Reward, and Conditional Guidance Design

  • Public
  • 384 runs
  • GitHub
  • Paper