Skip to content
AI-generated research idea🔴Hard💡Moderate InnovationRelated work found

Stratified Minute-Scale Rollout Distillation and Self-Resampled Base Stabilization for 60s+ Generation

Extend self-rollout distillation to minute-scale multi-event horizons via stratified late-window supervision and scheduled-sampling base tuning so both base and fast generators remain stable.

Time

a couple of months

Compute level

Large cluster

Resources

Large GPU cluster (64-128x 80GB+ GPUs consistent with MovieRoll Stage 2 and CuLT-DMD training) and 10K+ multi-event 60s prompt sequences.

What the paper used

Training used 32 to 128 GPUs in BF16 on an internal corpus (620K-1.3M clips, 5K prompts); inference benchmarks used 1 to 8 NVIDIA H20 GPUs.

Prerequisites

Distribution matching distillation (DMD), autoregressive self-rollout training, flow matching, and gradient accumulation/checkpointing at scale.

Scope

Extend CuLT-DMD training loop with stratified supervision and add a self-rollout base stabilization stage.

The Gap

In Section 5.2.1, Section 6.2, and Table 10/11, MovieRoll-Base is trained only up to 15s (361 frames) using WinTF with uniform random history noise perturbation U(0, 0.15), and CuLT-DMD Phase 2 trains only up to 30s (721 frames, 2 events) by accumulating local gradients across all K=9 supervision windows.

“Long-horizon stability also remains a limitation. MovieRoll-Base is insufficiently stable at 60s, so only MovieRoll-Distilled is evaluated at this duration.”
From the paper, Section 6.2, page 29. Matched against the PDF text.

Limitation

MovieRoll-Base collapses or becomes insufficiently stable to evaluate at 60s (1441 frames), and even MovieRoll-Distilled never trains on 60s rollouts or 3+ event transitions, leaving a train-test horizon gap beyond 30s.

Proposed Solution

Stabilize MovieRoll-Base by adding a Stage 3 scheduled-sampling self-rollout phase where the model generates 30s-60s trajectories using AnnealCFG and receives WinTF flow-matching supervision on late-horizon windows conditioned on self-generated history (instead of purely Gaussian-perturbed ground truth); simultaneously, extend CuLT-DMD to 60s (1441 frames, 3-6 events) using stratified window sampling (backpropagating on a random subset of K early, mid, and late supervision windows per rollout while detaching intermediate steps) to keep memory and compute bounded.

Why This Matters

Closing the training-to-inference horizon gap at 60s should make the higher-capacity multi-step base model usable for minute-scale generation and further curb late-horizon drift in the distilled generator.

How to Get Started

How to test it: Test whether self-rollout Stage 3 stabilization enables MovieRoll-Base evaluation at 60s and whether 60s stratified CuLT-DMD improves MovieRoll-Distilled's 1441-frame single- and 6-event HeliosBench TWR and Drifting Naturalness over the 721-frame-trained baseline; stop if 60s base Drifting Naturalness stays at 1 or 60s CuLT-DMD overfits/collapses dynamics.

  1. Evaluate the released MovieRoll-Base checkpoint at 60s (1441 frames) on 30 HeliosBench prompts to diagnose exact failure modes (e.g., saturation, blur, or repetition).
  2. Synthesize 10K 6-event 60s prompt sequences using the Streaming Video Prompt Rewriting Skill in Batch/Sequential mode.
  3. Implement stratified supervision window sampling in the CuLT-DMD training loop so 1441-frame rollouts backpropagate through 9 sampled early/mid/late windows with fixed per-step memory.
  4. Run a 300-step Stage 3 self-rollout stabilization run on MovieRoll-Base and a 600-step Phase 3 CuLT-DMD run on MovieRoll-Distilled, evaluating at 1441 frames on HeliosBench.

Suggested by AI. Adapt it to your setup.

Has This Been Done?

Related work found

Works like Self Forcing and End-to-End Training for Autoregressive Video Diffusion via Self-resampling address train-test gaps with self-rollout and resampling, but the extension of CuLT-DMD and scheduled-sampling distillation to 60-second generation using stratified window sampling is open.

We searched OpenAlex and arXiv for “minute-scale video diffusion self-rollout distillation exposure bias” and “Stratified Minute-Scale Rollout Distillation and Self-Resampled Base Stabilization for 60s+ Generation” and an AI compared the idea with the top 10 results. A small search can miss relevant work.

Source Paper

MovieRoll: A Systematic Framework for Rolling Out Long-Form Videos

Yunyang Ge, Weijie Kong, Ryan Xu +9 more · 2026

MovieRoll introduces a base-to-distilled framework for streaming long-form video generation that sustains visual quality, active motion, and prompt continuity up to 60 seconds while bounding historical KV-cache memory. By combining parallel rolling-window fine-tuning, hierarchical layer-wise cache compression, cumulative long-rollout distribution matching distillation, and streaming decoding infrastructure, it outperforms existing streaming baselines on single-event and multi-event HeliosBench evaluations.

Related Ideas

Is this a good idea?

Share onXLinkedIn