Where Predictive Supervision Goes Shapes What VLA Policies Learn
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
The Flow has not summarised this story yet — read it at arXiv AI.
arXiv:2609.23369v1 Announce Type: new Abstract: Generation-free world action models (WAMs) retain future-video prediction during training but act from internal video features at inference, leaving un...
arXiv:2609.37250v1 Announce Type: cross Abstract: World-action models (WAMs) couple future visual-state prediction with action generation. By adapting video generators or image-editing models pretrai...
The paper investigates how encouraging action predictability in latent world models can preserve intrinsic dynamics, which are crucial for planning. By deriving a theoretical lower bound and conducting empirical tests, the authors show that action predictability prevents the encoder from merging distinct action outcomes, even when visual appearance varies. Using a method called Concat IDM, they achieve a substantial increase in planning success on the Push‑T benchmark, from 9.1% to 77.2%, demonstrating the practical benefit of this inductive bias.
arXiv:2606. 15768v1 Announce Type: cross Abstract: Vision-Language-Action models (VLAs) leverage large-scale vision-language pretraining for semantic robot control, but often lack explicit foresight into how robot actions change the scene.
arXiv:2608. 11605v1 Announce Type: new Abstract: World Action Models (WAMs) couple future visual prediction with robot action generation, enabling policies to model how the physical world evolves during interaction.
SaiVLA-JEPA is a family of vision‑language‑action policies that condition on frozen V‑JEPA features and use lightweight auxiliary heads to predict future latent representations during training. Experiments on 10,030 LIBERO‑Plus episodes show that JEPA conditioning improves success rates to about 82.5% versus 71.6% without JEPA, but adding future supervision does not yield a detectable aggregate gain. Variants that adapt latent context or selected parameters provide mixed results, suggesting that latent‑prediction objectives should be evaluated via closed‑loop rollouts rather than latent loss alone.