AI News HubLIVE
原文

Balancing Multimodal Learning through Label Space Reshaping

Multimodal learning often suffers from modality imbalance, where faster-converging modalities dominate optimization. Existing methods typically strengthen weak modalities or adjust gradients, but may compromise strong modalities. This paper proposes Balanced Multimodal Label Reshaping (BMLR), the first label-side approach to promote balance. BMLR reshapes the cross-modal label space to equalize mapping difficulty across modalities, enhancing interaction and injecting rich inter-class information. Extensive experiments show consistent improvement and compatibility.

Article intelligence

EngineersAdvanced

Key points

  • Modality imbalance arises from differences in mapping difficulty from feature spaces to the shared label space.
  • BMLR is the first method to address multimodal balance from the label side.
  • BMLR reshapes the label space to equalize mapping difficulty, improving modality interaction.
  • Experiments across architectures demonstrate consistent performance gains and strong compatibility.

Why it matters

This matters because modality imbalance arises from differences in mapping difficulty from feature spaces to the shared label space.

Technical impact

May affect model selection, inference cost, product capability, and evaluation benchmarks.

[2605.28869] Balancing Multimodal Learning through Label Space Reshaping

[Submitted on 22 May 2026]

Title:Balancing Multimodal Learning through Label Space Reshaping

View a PDF of the paper titled Balancing Multimodal Learning through Label Space Reshaping, by Xiaoyu Ma and 5 other authors

View PDF HTML (experimental)

Abstract:Multimodal learning often suffers from modality imbalance, where modalities that converge faster dominate optimization while others remain undertrained. Existing approaches typically mitigate this issue by strengthening the weak modality or adjusting optimization gradients. However, such strategies mainly compensate for optimization rate discrepancies, often at the expense of the strong modality's optimization capacity, without analyzing how these discrepancies arise at the modality level. Based on theoretical insights and empirical observations, we argue that the discrepancy of learning pace arises from differences in the mapping difficulty between modality-specific feature space and the shared label space. To address this issue, we propose Balanced Multimodal Label Reshaping (BMLR), the first method that promotes multimodal balance from the label-side design. BMLR reshapes the cross-modal label space to equalize mapping difficulty across modalities, thereby facilitating modality interaction and injecting richer inter-class information into each modality. Extensive experiments across multiple architectures demonstrate that BMLR consistently improves multimodal performance and exhibits strong compatibility with diverse model designs. The source code will be released soon.

Comments: In process

Subjects:

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

Cite as: arXiv:2605.28869 [cs.LG]

(or arXiv:2605.28869v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2605.28869

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Xiaoyu Ma [view email] [v1] Fri, 22 May 2026 08:22:31 UTC (4,057 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Balancing Multimodal Learning through Label Space Reshaping, by Xiaoyu Ma and 5 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.LG

new | recent | 2026-05

Change to browse by:

cs cs.AI

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?)

IArxiv recommender toggle

IArxiv Recommender (What is IArxiv?)

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?)