PackForcing: Short Video Training Suffices for Long Video Sampling and Long Context Inference
Xiaofeng Mao, Shaohao Rui, Kaining Ying, Bo Zheng, Chuanhao Li +2 more
Abstract
Autoregressive video diffusion models have demonstrated remarkable progress, yet they remain bottlenecked by intractable linear KV-cache growth, temporal repetition, and compounding errors during long-video generation. To address these challenges, we present PackForcing, a unified framework that efficiently manages the generation history through a novel three-partition KV-cache strategy. Specifically, we categorize t...
Results & Benchmarks
| Task | Dataset | Metric | Value |
|---|---|---|---|
| Generation | CausVid yin2025slow | 0–10 s | 32.65 |
| Generation | LongLive yang2025longlive | 0–10 s | 33.95 |
Hardware Requirements
- Expect multi-day setup/compute for meaningful reproduction based on current guidance.
Best Implementation
Maintained implementation evidence is not confirmed for this paper yet.
Use the Implementation Status and Reproduction Path sections below for the current action plan.
Reproduction Path
Follow this baseline workflow to decide if this paper is worth immediate prototyping.
- 1
No maintained paper-verified implementation was found; start with the closest related repositories below.
- 2
Compare repo methods against the paper equations/algorithm before trusting metrics.
- 3
Create a minimal baseline implementation from the paper and use adjacent repos as references.
- 4
Prioritize reproducing the core method first: Diffusion.
Framework baselines
- Hugging Face Diffusers training guide
Practical baseline for diffusion model reproduction.
Related Implementations
These are not paper-verified. Use them as reference points when no direct implementation is available.
Matches contextual method/domain keyword: diffusion
Additional Implementations
No additional verified repositories beyond the primary recommendation.
Hugging Face Artifacts
No trustworthy direct or curated related Hugging Face artifacts were found yet.
Continue with targeted Hugging Face searches: