aiqbench analysis
This research introduces DILL, a framework designed to reduce brittleness in Vision-Language-Action models by preventing them from relying on spurious visual correlations. By enforcing domain-invariant latent representations through contrastive objectives and disentanglement regularization, the approach significantly improves robustness against distribution shifts while maintaining high task success rates. The findings offer critical insights for advancing reliable embodied AI systems that generalize effectively beyond controlled simulation environments to real-world manipulation tasks.
Disentangling Spurious Correlations in Vision-Language-Action Models via Predicting Domain-Invariant Latent Lookahead
arXiv:2609.37165v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models remain brittle under visual distribution shifts, often relying on spurious correlations tied to domain-specific factors rather than task-relevant structure. We propose Domain-Invariant Latent Lookahead (DILL), a representation-learning framework that mitigates shortcut learning in VLA policies. Our key idea is to supervise policies with domain-invariant future latents learned from domain-transformed trajectory data. A Task-Domain Encoder is trained with contrastive objectives and Gaussian disentanglement regularization to separate task-relevant structure from domain-specific visual variation. The learned encoder then provides future latents for VLA policy learning through lookahead prediction and domain disentanglement, encouraging the policy to focus on task-relevant structure rather than incidental visual factors. Counterfactual task-view evaluations show that DILL reduces shortcut reliance, while LIBERO-Plus evaluations demonstrate improved visual robustness, with 69.1% average success, 11.4 percentage points above the strongest baseline. Real-world manipulation experiments further support DILL's applicability beyond controlled simulation. Complementary latent-space diagnostics show that these behavioral gains are accompanied by representations that better preserve task-consistent structure while suppressing domain-specific variation. Our project page is available at https://dill-vla.github.io/.
Abstract:Vision-Language-Action (VLA) models remain brittle under visual distribution shifts, often relying on spurious correlations tied to domain-specific factors rather than task-relevant structure. We propose Domain-Invariant Latent Lookahead (DILL), a representation-learning framework that mitigates shortcut learning in VLA policies. Our key idea is to supervise policies with domain-invariant future latents learned from domain-transformed trajectory data. A Task-Domain Encoder is trained with contrastive objectives and Gaussian disentanglement regularization to separate task-relevant structure from domain-specific visual variation. The learned encoder then provides future latents for VLA policy learning through lookahead prediction and domain disentanglement, encouraging the policy to focus on task-relevant structure rather than incidental visual factors. Counterfactual task-view evaluations show that DILL reduces shortcut reliance, while LIBERO-Plus evaluations demonstrate improved visual robustness, with 69.1% average success, 11.4 percentage points above the strongest baseline. Real-world manipulation experiments further support DILL's applicability beyond controlled simulation. Complementary latent-space diagnostics show that these behavioral gains are accompanied by representations that better preserve task-consistent structure while suppressing domain-specific variation. Our project page is available at this https URL.
Accepted to CoRL 2026. Project website: this https URL
Robotics (cs.RO); Machine Learning (cs.LG)
arXiv:2609.37165 [cs.RO]
(or arXiv:2609.37165v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2609.37165 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 276 words
Related coverage
Concrete Evaluations to Investigate the OpenAI Model That Hacked Hugging Face
AI Alignment Forum · Aug 3, 2026
Embodied Interpretability: Linking Causal Understanding to Generalization in Vision-Language-Action Models
arXiv · Jun 11, 2026
Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · Sep 25, 2026
Partition-Aware Unlearning for Removing Spurious Correlations in Large Vision-Language Models
arXiv · Sep 1, 2026