Where this shows upLive preview

What it is

If a representation truly understands movement, it should be able to continue it. Predictive runs is our research direction for short-horizon forecasting: given the last seconds of skeletal motion, roll the model forward two to three seconds and read out likely continuations — runs, overlaps, pressing triggers — as probabilities, not certainties.

This page is labeled Research on purpose. Nothing here is shipped, and forecasting quality is a claim we will make with evaluation results — or not at all.

ExampleShort-horizon rollout: one observed trajectory, a fan of probable continuations. Conceptual illustration.

How it works

The mechanism under investigation is generative rollout: the same foundation model that fills gaps in observed movement and denoises it can, in principle, be sampled forward in time. Today the simulation stack is an evaluation workbench — reconstruction, denoising, controlled what-if probes — and forward rollouts are the research frontier built on top of it.

What it would make possible

If rollouts survive evaluation, review changes shape: analysts stop asking only what happened and start seeing what was likely to happen next — with the probabilities attached.

ParameterValueNotes
Horizon2–3 sShort enough to learn, long enough to matter.
Outputdistribution over continuationsProbabilities, never scripted predictions.
Readinganalyst-facing reviewDecision support in review, not in-game automation.
Gateevaluation results firstShips when forecast quality survives the evaluation suite.

Status & roadmap

Research

Research. Short-horizon forecasting is a direction, not a feature: it builds on validated reconstruction capability and inherits every caveat of proof-of-concept data scale. Progress will be published through the evaluation docs first.