proomt

Search

Search posts, papers, and topics

All posts

Hugging Face Daily PapersHaiyu Zhang, Wenqiang Sun, Tengfei Wang1 min readpaperadvanced

WorldPlay2: Extending Real-Time Interactive World Models in Control and Horizon

Read on arxiv.org

Summary

WorldPlay2 is an interactive world model designed for real-time response and long-horizon consistency. It achieves this through a factorized hybrid control interface and a co-design of compressed memory with stable distillation techniques.

  • WorldPlay2 uses a factorized hybrid control interface, combining frame-aligned action control with structured semantic control.
  • The semantic control explicitly disentangles scene appearance, character identity, and dynamic events for effective control learning.
  • Compressed memory tokens are used to efficiently store historical contexts, reducing distillation overhead for long-horizon modeling.
  • Stable Forcing initializes the autoregressive student via a few-step strategy and leverages full-rollout replay for robust distillation.

Researchers and practitioners in generative AI and simulation will find this relevant for developing more robust and efficient interactive world models with better control and long-term coherence.

7/10
machine learningcomputer visionworld modelsreal timedeep learninggenerative ai

Related reading

  1. Zing-0.5: Toward Playable Worlds with Real-Time Joint Action and Text Control

    Zing‑0.5 is a 5 B autoregressive world model that lets users control generated environments in real time using both keyboard actions and text prompts. The paper introduces unified action‑text conditioning, segment‑level teacher distillation, and a low‑cost streaming inference pipeline that runs at 24 FPS (832×480) for about $0.009 per minute, achieving 81 % overall and 88.5 % consistency on a nav…

    Hugging Face Daily Papersarxiv.org1 minpaper

  3. HelixWorld: A Real-time Interactive Audio-Visual World Model

    HelixWorld is a real-time interactive audio-visual world model that generates synchronized visual scenes and camera-grounded spatial stereo sound. It achieves drift-free joint audio-visual rollouts at 24 FPS on a single GPU, matching visual fidelity of silent models while significantly improving spatial-acoustic immersion.

    Hugging Face Daily Papersarxiv.org1 minpaper

  4. GameHorizon Suite: Multi-Horizon Data and Evaluation in Gameplay

    The paper presents GameHorizon, a unified suite comprising an automated annotation pipeline, a 5,000‑hour multi‑horizon gameplay dataset from 21 AAA titles, and reproducible offline and online benchmarks. Using it, the authors evaluate 47 models, exposing a hierarchy of task difficulty and gaps in long‑term planning.

    Hugging Face Daily Papersarxiv.org1 minpaper

  5. WorldLine: Action-Driven Visual Simulation for Robotic Manipulation

    WorldLine is an action-driven visual simulator for robotic manipulation that decouples transferable dynamics learning from heterogeneous action grounding. It improves prediction accuracy and task success by learning from vast action-free and action-conditioned robot video data, enabling more efficient policy evaluation and embodied planning.

    Hugging Face Daily Papersarxiv.org1 minpaper

  6. Ego2Act: Evaluating Goal-Directed Manipulation in Egocentric Video Generation

    Ego2Act introduces a new benchmark for evaluating goal-directed manipulation in egocentric video generation, featuring 2,640 videos across 110 real-world tasks. It reveals that current video generation models struggle with multi-step physical reasoning, often skipping steps and failing at fine-grained object manipulation and persistent world modeling.

    Hugging Face Daily Papersarxiv.org1 minpaper