{"ID":2922048,"CreatedAt":"2026-06-02T02:42:49.606572591Z","UpdatedAt":"2026-06-02T09:47:57.354342003Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.00658","arxiv_id":"2606.00658","title":"Collaborative Few-Step Distillation and Low-Bit Quantization for Wan2.2 Dual-Expert Video Diffusion Models","abstract":"Large video diffusion models achieve strong visual quality but remain expensive to deploy because each sample requires many denoising steps and a large resident parameter footprint. This paper studies a deployment-oriented compression pipeline for Wan2.2-T2V-A14B by combining few-step distribution-matching distillation with low-bit quantization. The pipeline follows the model's dual-expert denoising route, calibrates the high-noise and low-noise branches separately, protects sensitive entrance layers, and uses HiF4-style low-bit representation to improve dynamic-range coverage. Quantization is calibrated on the distilled few-step student rather than on the original long-step trajectory, reducing activation-distribution mismatch during inference. The proposed co-design keeps the quantized model close to the same-step full-precision model and surpasses the original full-precision baseline at 8 and 20 steps on average. The 20-step setting gives the best quality-efficiency trade-off in the tested configurations.","short_abstract":"Large video diffusion models achieve strong visual quality but remain expensive to deploy because each sample requires many denoising steps and a large resident parameter footprint. This paper studies a deployment-oriented compression pipeline for Wan2.2-T2V-A14B by combining few-step distribution-matching distillation...","url_abs":"https://arxiv.org/abs/2606.00658","url_pdf":"https://arxiv.org/pdf/2606.00658v1","authors":"[\"Jinyang Du\",\"Shenghao Jin\",\"Ziqian Xu\",\"Ruihao Gong\",\"Shiqiao Gu\",\"Yang Yong\",\"Jinyang Guo\",\"Xianglong Liu\"]","published":"2026-05-30T10:15:26Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Diffusion Model\"]","has_code":false}