AI News HubLIVE
Original source2 min read

Readable but Not Controllable: Neuron-Level Evidence for Medical LLM Hallucination

A new study shows that while medical LLM hallucination can be reliably detected (AUROC 0.77-0.86) using a simple probe, the underlying neural signal is distributed and redundant. Crucially, detection does not translate to control: steering hallucination-associated neurons fails to correct outputs, revealing a fundamental gap between readability and controllability.

SourcearXiv Computational LinguisticsAuthor: Vijay Vankadaru, Asha Matthews, Tanya Roosta, Peyman Passban

-->

[Submitted on 30 Jun 2026]

Title:Readable but Not Controllable: Neuron-Level Evidence for Medical LLM Hallucination

View a PDF of the paper titled Readable but Not Controllable: Neuron-Level Evidence for Medical LLM Hallucination, by Vijay Vankadaru and 3 other authors

View PDF HTML (experimental)

Abstract:Hallucination remains one of the central obstacles to deploying medical LLMs. Yet, even when hallucination can be detected, it is still unclear whether the internal representations associated with it can be used for control rather than detection alone. Using four open-source models across a suite of medical question-answering datasets, we show that a simple, carefully conditioned probe can reliably detect hallucination, with AUROC scores between 0.77 and 0.86 in our case. We further show that this signal is distributed and redundant rather than narrowly localized. Systematically selected neurons outperform random neurons only at very small subset sizes, whereas random subsets of a few hundred neurons recover nearly the full signal, and low-dimensional random projections preserve most of the detection performance. Beyond detection, we test whether this representation is causally actionable. Across 16 model--dataset combinations, our results reveal a sharp gap between decodability and controllability. The same internal structure that makes hallucination easy to detect does not translate into reliable neuron-level control. These findings show that medical hallucination seems to be readily visible in internal activations, but not easily corrected by steering the neurons most associated with it. More broadly, our results suggest that hallucination mitigation is not simply a matter of identifying the right neurons, and point to a deeper separation between what representations reveal and what they allow us to change.

Subjects:

Computation and Language (cs.CL)

Cite as: arXiv:2607.00158 [cs.CL]

(or arXiv:2607.00158v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2607.00158

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Peyman Passban [view email] [v1] Tue, 30 Jun 2026 20:34:45 UTC (445 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Readable but Not Controllable: Neuron-Level Evidence for Medical LLM Hallucination, by Vijay Vankadaru and 3 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-07

Change to browse by:

cs

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?)