[2602.04105] Expert Selections In MoE Models Reveal (Almost) As Much As Text
-->
Computer Science > Computation and Language
arXiv:2602.04105 (cs)
[Submitted on 4 Feb 2026 (v1), last revised 13 Mar 2026 (this version, v3)]
Title:Expert Selections In MoE Models Reveal (Almost) As Much As Text
Authors:Amir Nuriyev, Gabriel Kulp<br>View a PDF of the paper titled Expert Selections In MoE Models Reveal (Almost) As Much As Text, by Amir Nuriyev and 1 other authors
View PDF
Abstract:We present a text-reconstruction attack on mixture-of-experts (MoE) language models that recovers tokens from expert selections alone. In MoE models, each token is routed to a subset of expert subnetworks; we show these routing decisions leak substantially more information than previously understood. Prior work using logistic regression achieves limited reconstruction; we show that a 3-layer MLP improves this to 63.1% top-1 accuracy, and that a transformer-based sequence decoder recovers 91.2% of tokens top-1 (94.8% top-10) on 32-token sequences from OpenWebText after training on 100M tokens. These results connect MoE routing to the broader literature on embedding inversion. We outline practical leakage scenarios (e.g., distributed inference and side channels) and show that adding noise reduces but does not eliminate reconstruction. Our findings suggest that expert selections in MoE deployments should be treated as sensitive as the underlying text.
Subjects:
Computation and Language (cs.CL); Cryptography and Security (cs.CR)
Cite as:<br>arXiv:2602.04105 [cs.CL]
(or<br>arXiv:2602.04105v3 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2602.04105
Focus to learn more
arXiv-issued DOI via DataCite
Submission history<br>From: Amir Nuriyev [view email]<br>[v1]<br>Wed, 4 Feb 2026 00:42:30 UTC (366 KB)
[v2]<br>Thu, 12 Mar 2026 17:39:30 UTC (367 KB)
[v3]<br>Fri, 13 Mar 2026 06:37:48 UTC (367 KB)
Full-text links:<br>Access Paper:
View a PDF of the paper titled Expert Selections In MoE Models Reveal (Almost) As Much As Text, by Amir Nuriyev and 1 other authors<br>View PDF<br>TeX Source
view license
Current browse context:
cs.CL
next >
new<br>recent<br>| 2026-02
Change to browse by:
cs<br>cs.CR
References & Citations
NASA ADS<br>Google Scholar
Semantic Scholar
export BibTeX citation<br>Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
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? |<br>Disable MathJax (What is MathJax?)