Emergent Compositional Skills in Mixture-of-Experts VLAs
This research explores using a VLA with a Mixture-of-Experts (MoE) action head to learn compositional robot policies end-to-end from demonstrations, without predefined task decomposition. The experts are reused across tasks and correspond to distinct low-level behaviors, with the router implicitly performing high-level sequencing, achieving performance on par with monolithic baselines while demonstrating interpretable specialization.
-->
[Submitted on 22 Jul 2026]
Title:Emergent Compositional Skills in Mixture-of-Experts VLAs
View a PDF of the paper titled Emergent Compositional Skills in Mixture-of-Experts VLAs, by Shlok Shah and 4 other authors
View PDF HTML (experimental)
Abstract:We consider the problem of learning compositional robot policies end-to-end from expert demonstrations, without any pre-specified notion of task decomposition or hierarchy. We ask whether a VLA trained with a simplified Mixture-of-Experts (MoE) action head can emergently learn to decompose tasks into reusable, interpretable primitives. We find that learned experts are heavily reused across tasks and consistently correspond to qualitatively distinct low-level behaviors, suggesting that the router implicitly learns to perform high-level sequencing while experts serve as compositional primitives. Our MoE matches the task performance of a monolithic baseline while demonstrating meaningful expert specialization, a step toward modular, interpretable robot policies that emerge from data alone.
Comments: Accepted to the 2nd Workshop on Compositional Learning at ICML 2026
Subjects:
Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Cite as: arXiv:2607.20771 [cs.RO]
(or arXiv:2607.20771v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2607.20771
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Chirayu Nimonkar [view email] [v1] Wed, 22 Jul 2026 22:36:52 UTC (6,946 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Emergent Compositional Skills in Mixture-of-Experts VLAs, by Shlok Shah and 4 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.RO
new | recent | 2026-07
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?)