ActTime: Action-Conditioned World Modeling for Time Series
We introduce ActTime, an action-conditioned world model for time series.
ProofPaper ↗
Key points
- Time-series models increasingly condition on known future inputs.
- In controlled systems, some of these inputs are actions that can be changed, compared, and optimized before execution.
- Over the observed history, ActTime updates variable-specific latent states through an action-conditioned transition and corrects them using observations.
- We construct a simulator-grounded evaluation suite across building energy, traffic, greenhouse climate, and inventory management, covering factual prediction, intervention-effect prediction, offline decision making, and closed-loop planning.
Sources (1)
- [1]ActTime: Action-Conditioned World Modeling for Time SeriesarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 19, 04:24 AM
We introduce ActTime, an action-conditioned world model for time series.
Time-series models increasingly condition on known future inputs.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 19, 2026Vector-Quantized World Model: Latent Feedback Motion Planning for Robot Manipulation
- Oct 19, 2026Ropedia W0: Scaling Loco-Manipulation World Models across Embodiments with Human Experience
- Oct 19, 2026ActionRoPE: action as coordinates for spatially consistent control in planar-camera game world models
- Oct 16, 2026Beyond Large Language Models: A Mental World Model Framework for Mental Modeling, Simulation, Evolution, and Interpretability
- Oct 16, 2026Momentum: Persistent Dynamics for Video World Models
- Oct 16, 2026ORD-WM: Learning Object-Centric Rigid Dynamics for 3D World Modeling in Robotic Manipulation