4LC 4D latent cache makes world-model conditioning 68.5x faster with less GPU memory
ZGCA · hf · 2026-10-14
Researchers from Shanghai Jiao Tong University and the Chinese Academy of Sciences introduce 4LC, a 4D latent cache for efficient conditioning in dynamic world modeling for video generation.
- Reuses visual representations of static and dynamic content directly as generation conditions, avoiding per-request RGB rendering and re-encoding
- Composite cache binds representations from multiple encoders (VAE, DINO, V-JEPA, VGGT) to shared geometry and motion for richer descriptions
- Flexible spatiotemporal interface reads the cache at independently specified times and viewpoints, combining readouts across observations
- 4LC-Video handles camera control, temporal control, novel-view synthesis, scene revisits, and long-term generation with a single model, no task-specific fine-tuning
- Per the paper's figures: latency 40.14ms vs. 2750ms baselines (68.5x faster) and 0.385GiB GPU memory (12.6x lower)
More from Research
- FlashVSR+ runs untiled 4K video super-resolution on one RTX 4090 with 2x speedup — MMLab-CUHK · 2026-10-14
- Sber's GreenWAM world foundation model spans 20 robot families as both predictor and policy — SberRoboticsCenter · 2026-10-14
- Tencent's LifeSearch benchmark: best deep research agent scores just 63.4% on real-life tasks — Tencent-Hunyuan · 2026-10-14
- Science: Menopause may explain why women face higher dementia risk than men — AnnaCiaunica · 2026-10-14
- When will AI start delivering breakthroughs in physics like it has in math? — burny_tech · 2026-10-14
- Photonic Training of a 1B GPT-2 Without Backprop Lands in PNAS — IgorCarron · 2026-10-14