AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Improving Multimodal Learning via Imbalanced Learning

Shicai Wei, Chunbo Luo, Yang Luo

ICCV 2025 · Vision-language and multimodal › Cross-modal learning · approach: Multimodal fusion with missing modalities · 19 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add unimodal bias regularization (ablation)ARL · CREMA-Daccuracy (%): 58.83 -> 67.20Improved
add asymmetric learning (ablation)ARL · CREMA-Daccuracy (%): 58.83 -> 70.23Improved
combine UR and AL (ablation)ARL · CREMA-Daccuracy (%): 70.23 -> 74.45Improved
add gradient residual (ablation)ARL · CREMA-Daccuracy (%): 74.45 -> 76.61Improved
upweight the better modality's gradient fivefold (imbalance enlarged)multimodal baseline · CREMA-DaccuracyImproved
apply balance-based modulation (OGM-GE, AGM, PMR) instead of concatenationmultimodal baseline · AVEaccuracy (%): 66.15 -> 65.62Hurt
apply ARL on AVE vs concatenationmultimodal baseline · AVEaccuracy (%): 66.15 -> 72.89Improved
add ARL to MMTM intermediate fusionMMTM · CREMA-Daccuracy (%): 51.86 -> 58.68Improved
add ARL to mmFormermmFormer · CREMA-Daccuracy (%): 59.69 -> 68.33Improved
add ARL to Swin-Base with gated fusionSwin-Base · AVEaccuracy (%): 84.52 -> 91.11Improved
use ARL instead of best prior method (D&R)UCF101 · UCF101accuracy (%): 82.11 -> 83.22Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy