Skip to content
AI News HubLIVE
Original source2 min read

VG-Steer: Visual-Geometric Policy Steering from In-Context Aligned Abstractions

Summary

VG-Steer is a test-time adaptation method that corrects frozen robot policies by transferring nuanced motion from successful in-distribution rollouts, captured along object-relational distance pairs, using large vision and vision-language models. In simulation, it efficiently corrects flow-matching policies under pose and task-goal shifts; on a real robot, it more than doubles success rate on cable placements one grid step outside training positions and steers a single-task policy toward unseen objects while nearly eliminating object confusion.

SourcearXiv RoboticsAuthor: Arthur Bucker, Pramod Anantharam, Jonathan Francis, Jean Oh
VG-Steer: Visual-Geometric Policy Steering from In-Context Aligned Abstractions
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 18 Nov 2026]

Title:VG-Steer: Visual-Geometric Policy Steering from In-Context Aligned Abstractions

View a PDF of the paper titled VG-Steer: Visual-Geometric Policy Steering from In-Context Aligned Abstractions, by Arthur Bucker and 2 other authors

View PDF HTML (experimental)

Abstract:Adapting robot policies to out-of-distribution (OOD) settings with limited real-world data remains a key bottleneck to generalist robots. The multimodality and scarcity of robotics data lead end-to-end approaches to overfit and break under small distribution shifts. Test-time adaptation can alleviate this by correcting frozen policies at inference, but most methods ground their correction in a signal learned implicitly from robotic datasets, making it susceptible to the same shifts as the policy it corrects. Recent methods that instead exploit explicit spatial information are typically limited to coarse corrections, missing the nuanced, task-relevant motion most manipulation tasks require. We propose VG-Steer, which corrects existing policies by transferring the nuanced motion of successful in-distribution rollouts, captured along meaningful geometric axes, known here as object-relational distance pairs. VG-Steer leverages the semantic understanding of Large Vision Models (LVMs) and Vision-Language Models (VLMs) reasoning, which combines the task-relevant entity pairs and their object-relational distance projections into a single tailored visual-geometric guidance signal. In a simulation benchmark, it efficiently corrects flow-matching policies under pose and task-goal shifts; on a real robot experiment, it widens the region of the workspace the frozen policy covers, more than doubling its success rate on cable placements one grid step outside the training positions, and steers a single-task policy toward objects it was never trained on while nearly eliminating object confusion. Project website: this https URL

Subjects:

Robotics (cs.RO); Artificial Intelligence (cs.AI)

Cite as: arXiv:2611.25199 [cs.RO]

(or arXiv:2611.25199v1 [cs.RO] for this version)

https://doi.org/10.48550/arXiv.2611.25199

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Arthur Fender Coelho Bucker [view email] [v1] Wed, 18 Nov 2026 21:18:04 UTC (11,811 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled VG-Steer: Visual-Geometric Policy Steering from In-Context Aligned Abstractions, by Arthur Bucker and 2 other authors

View PDF

HTML (experimental)

TeX Source

view license

Additional Features

Audio Summary

Current browse context:

cs.RO

new | recent | 2026-11

Change to browse by:

cs cs.AI

References & Citations

SciX

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

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.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • VG-Steer corrects frozen policies by transferring nuanced motion from successful in-distribution rollouts along object-relational distance pairs, avoiding reliance on implicitly learned signals that may suffer from distribution shifts.
  • The method leverages LVMs and VLMs to combine task-relevant entity pairs and their object-relational distance projections into a tailored visual-geometric guidance signal.
  • In simulation, VG-Steer efficiently corrects flow-matching policies under pose and task-goal shifts.
  • On a real robot, it widens the workspace coverage of a frozen policy, more than doubling success rate on cable placements one grid step outside training positions, and steers a single-task policy toward unseen objects while nearly eliminating object confusion.

Highlights and analysis are generated automatically and may contain errors. Check the original source.