ActionRoPE encodes action as world coordinates for spatially consistent game world models
Zhiyang Deng · hf · 2026-10-20
Tencent, NUS, and NTU researchers propose ActionRoPE for spatially consistent control in planar-camera game world models.
- Encodes action-driven camera motion into world-coordinate rotary position embeddings (RoPE), so the same landmark keeps a consistent token index before and after camera moves
- Pairs with coordinate-conditioned noise scheduling to stabilize observed regions before generating new content
- Supports arbitrary camera speeds, camera rotation, cross-scene generalization, and physically consistent motion with scene revisits
More from Multimodal
- Feeding the whole Animator's Survival Kit to Opus 5.5 yields decent keyframe animation — MoonL88537 · 2026-10-20
- Glimpse launches as a visual thinking model claiming Astra-level reasoning benchmarks — LinusEkenstam · 2026-10-20
- ComfyUI lets you pipe images into an LLM — how do you do the same with audio? — PhilosopherSweaty826 · 2026-10-20
- AI-generated mecha battle video looks straight out of a movie — Key_Layer6250 · 2026-10-20
- Free ComfyUI templates chain local LLM lyrics into YuE2 song rendering, runs on 12GB VRAM — Vivid_Promise1700 · 2026-10-20
- DeepSeek 4.1 Flash builds video animation in 12 min for $0.20, self-scores 7.5/10 — oran_ge · 2026-10-20