WorldBridge Boosts Task Generalization in World Action Models
WorldBridge separates video and action experts to achieve state-of-the-art results on RoboTwin 2.0 and other benchmarks.
Stories per day · 30 days
2 RSS
6 stories, newest first · one card per story
Stories tagged #video-prediction
RynnWorld-Latent conditions a world model on universal latent actions
Pretraining spans 23 datasets and about 553.5 million frames, then adapts to target robots without a shared action space.
DreamTrue: Action-Faithful Robot World Model with Counterfactual Post-Training
DreamTrue cuts AgiBot robot interaction defects from 48.12% to 6.25% and wins the 2026 world-model track.
Latent-Foresight: End-to-End Learning Predictable Representations for Latent World Models
Latent-Foresight jointly learns a tokenizer and flow model so vision latents stay predictable.
MinkowskiPE: Minkowski Positional Encoding for Spatiotemporal Perception
MinkowskiPE encodes relative spacetime displacement in attention and improves molecular and video prediction.
Frozen Flows Forget: Diagnosing and Restoring Lost Motion in a Latent-flow World Model
Latent-flow world models lose object motion; decode-augmented training restores it without unfreezing the representation.