aiqbench analysis
This research introduces a framework for robotic disassembly that leverages vision-language models to ground language instructions in spatial visual representations, significantly improving task success rates across diverse configurations. By combining hierarchical selection with context-aware imitation learning, the method effectively handles long-horizon manipulation sequences without requiring explicit object annotations. The approach demonstrates substantial performance gains over baseline diffusion policies and previous context-aware baselines in complex assembly environments.
Vision-Language Grounded Task-Context-Aware Imitation Learning for Robotic Disassembly
arXiv:2609.17714v1 Announce Type: new Abstract: Real-world robotic disassembly requires long-horizon execution, where robots must perform ordered sequences of manipulation tasks across multiple parts within a single scene. Multiple valid task goals and diverse assembly configurations make it difficult for imitation policies to infer the intended skill from raw observations alone, particularly when training data cannot cover the combinatorial diversity of real-world configurations and part geometries. We show that incorporating task context through language alleviates these challenges by providing explicit structure for skill selection and associating language-specified tasks with their corresponding manipulation targets in the visual scene. The proposed framework combines hierarchical task selection with task-context-aware imitation learning to ground language instructions in spatial visual representations for robotic disassembly. The resulting framework generalizes across diverse connector geometries and assembly configurations without requiring explicit object annotations. Our method improves end-to-end task success by 35 percentage points over the baseline diffusion policy and by 75 percentage points over the previous task-context-aware baseline.
Abstract:Real-world robotic disassembly requires long-horizon execution, where robots must perform ordered sequences of manipulation tasks across multiple parts within a single scene. Multiple valid task goals and diverse assembly configurations make it difficult for imitation policies to infer the intended skill from raw observations alone, particularly when training data cannot cover the combinatorial diversity of real-world configurations and part geometries. We show that incorporating task context through language alleviates these challenges by providing explicit structure for skill selection and associating language-specified tasks with their corresponding manipulation targets in the visual scene. The proposed framework combines hierarchical task selection with task-context-aware imitation learning to ground language instructions in spatial visual representations for robotic disassembly. The resulting framework generalizes across diverse connector geometries and assembly configurations without requiring explicit object annotations. Our method improves end-to-end task success by 35 percentage points over the baseline diffusion policy and by 75 percentage points over the previous task-context-aware baseline.
Accepted for publication in IEEE Robotics and Automation Letters (RA-L), 2026. 8 figures
Robotics (cs.RO)
arXiv:2609.17714 [cs.RO]
(or arXiv:2609.17714v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2609.17714 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
https://doi.org/10.1109/LRA.2026.3734885 Focus to learn more DOI(s) linking to related resources
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 · 270 words
Related coverage
What's Continual Learning, and Why Might We Expect To See It In Advanced LLM Agents?
LessWrong · Jun 12, 2026
DriveJudge: Rethinking Autonomous Driving Evaluation with Vision-Language Models
arXiv · Jun 17, 2026
Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy — Jason Ma, Dyna Robotics
Latent Space · Oct 13, 2026
[AINews] not much happened today
Latent Space · Sep 10, 2026