aiqbench analysis
ContextFlow advances robotics by integrating continuous flow-matching models with in-context learning to overcome the discretization errors inherent in autoregressive next-token prediction. By compressing multimodal demonstrations into latent representations, this method enables generalization to unseen task configurations without requiring fine-tuning on new tasks. The approach demonstrates significant gains in success rates for both simulated and real-world robotic manipulation scenarios.
ContextFlow: In-Context Flow Matching for Robot Manipulation
arXiv:2609.06852v1 Announce Type: new Abstract: Although highly effective in vision and language domains, applying in-context learning to robotics remains challenging. Existing autoregressive in-context imitation methods discretize continuous actions and exacerbate the accumulation of early prediction errors through next-token prediction, limiting their generalization on unseen task configurations. Meanwhile, flow-matching policies have been explored for continuous robot control and can help mitigate compounding errors; however, in-context imitation learning within a flow-matching framework remains underexplored. To address these limitations, we introduce ContextFlow, a conditional flow-matching model that learns continuous action distributions for in-context imitation learning. ContextFlow conditions flow-based action prediction on demonstrations and observations, enabling robust generation from noisy action distributions. To better encode multimodal in-context demonstrations, we adapt perceiver-style multimodal context compressors that distill visual, proprioceptive, and action sequences into compact, task-relevant latent representations. On LIBERO, ContextFlow outperforms ICRT by 35 percentage points in average success rate on unseen task configurations, while matching the performance of the task-specific fine-tuned VLA model $\pi_0$ without any fine-tuning on unseen tasks. On real robots, it generalizes to unseen configurations of both single-arm and bimanual tasks, achieving 40% success on a new pen-uncapping configuration. Project Page: https://dingjiansw101.github.io/contextflow-page/.
Abstract:Although highly effective in vision and language domains, applying in-context learning to robotics remains challenging. Existing autoregressive in-context imitation methods discretize continuous actions and exacerbate the accumulation of early prediction errors through next-token prediction, limiting their generalization on unseen task configurations. Meanwhile, flow-matching policies have been explored for continuous robot control and can help mitigate compounding errors; however, in-context imitation learning within a flow-matching framework remains underexplored. To address these limitations, we introduce ContextFlow, a conditional flow-matching model that learns continuous action distributions for in-context imitation learning. ContextFlow conditions flow-based action prediction on demonstrations and observations, enabling robust generation from noisy action distributions. To better encode multimodal in-context demonstrations, we adapt perceiver-style multimodal context compressors that distill visual, proprioceptive, and action sequences into compact, task-relevant latent representations. On LIBERO, ContextFlow outperforms ICRT by 35 percentage points in average success rate on unseen task configurations, while matching the performance of the task-specific fine-tuned VLA model $\pi_0$ without any fine-tuning on unseen tasks. On real robots, it generalizes to unseen configurations of both single-arm and bimanual tasks, achieving 40% success on a new pen-uncapping configuration. Project Page: this https URL.
Robotics (cs.RO)
arXiv:2609.06852 [cs.RO]
(or arXiv:2609.06852v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2609.06852 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 · 281 words
Related coverage
Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy — Jason Ma, Dyna Robotics
Latent Space · Oct 13, 2026
RedFlow: Redirect Failure into Action-Level Corrections for Flow-matching VLA Policy
arXiv · Jul 31, 2026
PRISM: Precision Forcing for Generalist Vision-Language-Action Policies
arXiv · Oct 14, 2026
KAN We Flow? Advancing Robotic Manipulation with 3D Flow Matching via KAN & RWKV
arXiv · Aug 19, 2026