Skip to content
AI News HubLIVE
Original source2 min read

See Less, Specify More: Visual Evidence Budgets for Generalizable VLAs

Summary

The paper introduces S2, a framework that improves vision-language-action model generalization by refining language supervision and imposing an explicit visual evidence budget. On eight real-robot tasks, mean subtask success rises from 54.2% to 79.0% over pi0.5.

SourcearXiv RoboticsAuthor: Yueh-Hua Wu, Tatsuya Matsushima, Kei Ota
See Less, Specify More: Visual Evidence Budgets for Generalizable VLAs
Report an error

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

Correction instructions
Read article

[2606.02735] See Less, Specify More: Visual Evidence Budgets for Generalizable VLAs

[Submitted on 1 Jun 2026]

Title:See Less, Specify More: Visual Evidence Budgets for Generalizable VLAs

View a PDF of the paper titled See Less, Specify More: Visual Evidence Budgets for Generalizable VLAs, by Yueh-Hua Wu and 2 other authors

View PDF HTML (experimental)

Abstract:Generalization remains a central bottleneck for vision-language-action (VLA) models: under distractors, appearance shifts, and semantically similar tasks, the policy must often infer local execution details from coarse instructions while also deciding which parts of the image matter for control. We present S2 (See Less, Specify More), a framework for improving VLA generalization by training the executor under a cleaner interface.

Specify More preserves the original instruction as a stable high-level goal while relabeling each trajectory into refined trajectory- and subtask-level language that disambiguates the current execution mode. Unlike native attention, See Less imposes an explicit visual evidence budget, training the executor to act from task-sufficient evidence rather than unconstrained visual context, without any region or mask annotation.

This interface lets the executor follow detailed guidance without relying on distracting visual patches or resolving avoidable ambiguity on its own, and it remains compatible with off-the-shelf VLM planners through in-context learning. Across our main evaluation settings, S2 improves overall generalization metrics by changing the executor's learning problem: coarse instructions induce avoidable supervision aliasing, goal-preserving local guidance outperforms instruction replacement in our main ablations, and explicit evidence budgeting reduces dependence on broad visual context beyond efficiency considerations.

Across eight real-robot tasks on TX-G2 (an AgiBot G2-compatible variant) and HSR, S2 raises mean subtask success from 54.2% to 79.0% over pi0.5. Together, these results suggest that VLA generalization improves when the executor is trained to act from informative local guidance and task-sufficient visual evidence, rather than recovering both from weak supervision.

Comments: Project page: this https URL

Subjects:

Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)

Cite as: arXiv:2606.02735 [cs.RO]

(or arXiv:2606.02735v1 [cs.RO] for this version)

https://doi.org/10.48550/arXiv.2606.02735

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Yueh-Hua Wu [view email] [v1] Mon, 1 Jun 2026 18:02:07 UTC (9,658 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled See Less, Specify More: Visual Evidence Budgets for Generalizable VLAs, by Yueh-Hua Wu and 2 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.RO

new | recent | 2026-06

Change to browse by:

cs cs.AI cs.LG

References & Citations

NASA ADS

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

InvestorsAdvanced

Key points

  • S2 relabels trajectories with refined trajectory- and subtask-level language while preserving the original instruction as a high-level goal.
  • An explicit visual evidence budget trains the executor to act from task-sufficient evidence without region or mask annotations.
  • Across eight real-robot tasks on TX-G2 and HSR, S2 improves mean subtask success from 54.2% to 79.0% compared to pi0.5.

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