Ruihao Gong

The paper introduces an efficient, novel algorithm for incremental Byte Pair Encoding (BPE) tokenization that processes input text prefix by prefix, achieving significant speedups and enabling streaming output.

The paper proposes a compression pipeline combining few-step distillation and low-bit quantization to significantly reduce the deployment cost and parameter footprint of large dual-expert video diffusion models like Wan2.2, while maintaining high visual quality.