aiqbench analysis
This paper introduces SlotFlow, a flow matching model for robot manipulation that enhances robustness by decoupling semantic object features from spatial cues. By suppressing irrelevant background correlations while retaining useful historical motion priors, the approach effectively handles visual distractors and spatial perturbations without sacrificing inference efficiency. This advancement addresses critical limitations in current autoregressive and diffusion-based policies regarding out-of-distribution shifts.
Object-Centric Conditioning for Visuomotor Flow Matching
arXiv:2609.24155v1 Announce Type: new Abstract: Robot visuomotor policies are commonly formulated as autoregressive, diffusion-based, or more recently, flow matching models. Among them, Action-to-Action (A2A) flow matching improves inference efficiency by initializing generation from historical action priors rather than stochastic noise. However, stale historical motion patterns and entangled global visual representations can jointly reduce robustness under spatial out-of-distribution (OOD) shifts and visual distractors. In this work, we propose SlotFlow, an object-centric flow matching policy for robust visuomotor manipulation. SlotFlow decouples scene observations into semantic ("what") features and lightweight image-plane spatial ("where") cues to provide object-aware policy conditioning and current-state grounding. The semantic representation suppresses irrelevant background correlations, while the spatial cue improves adaptation to shifted object configurations. Extensive simulation and real-world experiments demonstrate improved robustness under visual distractors and severe spatial perturbations while preserving the low-step inference efficiency of A2A. Controlled initialization and perception ablations further identify object-centric grounding as a major source of the gains and show that it complements, rather than replaces, useful historical motion priors.
Abstract:Robot visuomotor policies are commonly formulated as autoregressive, diffusion-based, or more recently, flow matching models. Among them, Action-to-Action (A2A) flow matching improves inference efficiency by initializing generation from historical action priors rather than stochastic noise. However, stale historical motion patterns and entangled global visual representations can jointly reduce robustness under spatial out-of-distribution (OOD) shifts and visual distractors. In this work, we propose SlotFlow, an object-centric flow matching policy for robust visuomotor manipulation. SlotFlow decouples scene observations into semantic ("what") features and lightweight image-plane spatial ("where") cues to provide object-aware policy conditioning and current-state grounding. The semantic representation suppresses irrelevant background correlations, while the spatial cue improves adaptation to shifted object configurations. Extensive simulation and real-world experiments demonstrate improved robustness under visual distractors and severe spatial perturbations while preserving the low-step inference efficiency of A2A. Controlled initialization and perception ablations further identify object-centric grounding as a major source of the gains and show that it complements, rather than replaces, useful historical motion priors.
Robotics (cs.RO)
arXiv:2609.24155 [cs.RO]
(or arXiv:2609.24155v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2609.24155 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 · 255 words
Related coverage
Faster Visuomotor Policy Learning on Action Manifolds via Riemannian MeanFlow
arXiv · Sep 25, 2026
OccWAM: Occupancy World-Action Models with Joint Scene-to-Ego Flow Matching
arXiv · Oct 14, 2026
COMiT: Learning Structured Visual Tokens through Sequential Communication
arXiv · Sep 30, 2026
Test-time Adversarial Takeover: A Real-time Hijacking Interface against Robotic Diffusion Policies
arXiv · Jun 10, 2026