2026-06-17原文2 min readUpdated: 2026-06-17

Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs

A new study identifies 'editing decoupling failure' in Multimodal Large Language Models: knowledge updates succeed under multimodal inputs but revert when inputs are split into unimodal. The authors propose DECODE, which disentangles and localizes modality-specific neuron groups for consistent knowledge editing across modalities.

SourcearXiv Machine LearningAuthor: Tingchao Fu, Wenkai Wang, Fanxiao Li, Huadong Zhang, Jinhong Zhang, Dayang Li, Yunyun Dong, Renyang Liu, Wei Zhou

[2606.17057] Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs

[Submitted on 20 Apr 2026]

Title:Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs

View a PDF of the paper titled Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs, by Tingchao Fu and 8 other authors

View PDF HTML (experimental)

Abstract:Although Knowledge Editing provides an efficient mechanism for updating the knowledge of Multimodal Large Language Models (MLLMs), we find that current paradigms still suffer from an important yet remain underexplored issue : editing decoupling failure, where entity-related knowledge can be updated when the model is triggered by multimodal inputs (text--image query pairs), however, it often reverts to outdated pre-edit facts when the paired inputs are split into unimodal ones. Our in-depth empirical analysis reveals that the entity knowledge in MLLMs is not stored as a unified representation, but is instead distributed across disentangled modality-specific pathways. As a result, updates biased toward multimodal queries fail to propagate effectively to unimodal circuits. To bridge this gap, we propose DECODE, which explicitly disentangles and localizes modality-specific neuron groups for targeted knowledge. Extensive experiments demonstrate that DECODE consistently achieves effective knowledge updates under different modality triggers, thereby mitigating editing decoupling failures.

Comments: 18 pages, 11 figures

Subjects:

Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)

Cite as: arXiv:2606.17057 [cs.LG]

(or arXiv:2606.17057v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2606.17057

arXiv-issued DOI via DataCite

Submission history

From: Tingchao Fu [view email] [v1] Mon, 20 Apr 2026 05:43:19 UTC (1,631 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs, by Tingchao Fu and 8 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.LG

new | recent | 2026-06

Change to browse by:

cs cs.AI cs.CL

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

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

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)