WorldCast: A Distributed Multiplayer World Model That Runs Counter-Strike 2 in Real Time
CUHKSZ · hf · 2026-10-14
Researchers from CUHK-Shenzhen, Tsinghua, Didi Voyager and others present WorldCast, a distributed multiplayer world model where each player runs a local client with a video generator and a state model, exchanging only player and scene states between clients.
Key points:
- The state model estimates each player's position from generated video and controls, trained with recorded positions and map geometry.
- Camera-aligned player state fields improve player rendering rates by over an order of magnitude vs joint multi-view generation.
- Shared scene state keeps views consistent across whole rounds; on Counter-Strike 2 the system runs in real time, scales without centralized compute, and keeps image quality stable over hour-long rollouts.
More from Research
- AIPOCH's open-source Open-Science workbench: reproducible AI research sessions — ahuja_priyank · 2026-10-14
- AIPOCH's Open-Science ships portable .science packages that share full research context, not just outputs — ahuja_priyank · 2026-10-14
- AIPOCH's Open-Science packages the full path to AI-assisted research results into .science files — ahuja_priyank · 2026-10-14
- Video-Arena: Game-Based Video Benchmark Shows Native Models Score Just 13.2 F1 — Kuan Zhang · 2026-10-14
- LSCBench: Vibe Coding Agents Write Functional Code, But Only 25% Is Also Secure — Qwen · 2026-10-14
- Omni-LLMs Know More Than They Say: NAVER's OmniCatalyst Adds 7.29% via Self-Generated Text — naver-ai · 2026-10-14