AI Paper Atlas

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

Provable Dynamic Fusion for Low-Quality Multimodal Data

Qingyang Zhang, Haitao Wu, Changqing Zhang, Qinghua Hu, Huazhu Fu, Joey Tianyi Zhou, Xi Peng

ICML 2023 · Vision-language and multimodal › Cross-modal learning · approach: Multimodal fusion with missing modalities · 20 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
Use QMF instead of TMCmultimodal classification · NYU Depth V2, Gaussian noise ε=10.0average accuracy: 53.36 -> 55.60Improved
Use QMF instead of Late fusionmultimodal classification · SUN RGB-D, Gaussian noise ε=10.0average accuracy: 47.33 -> 48.58Improved
Use QMF instead of TMCmultimodal classification · FOOD-101, Gaussian noise ε=10.0average accuracy: 61.37 -> 62.21Improved
Use QMF instead of TMCmultimodal classification · MVSA, Gaussian noise ε=10.0average accuracy: 60.36 -> 61.28Improved
Add regularization term (ablation)QMF without UAW or regularization · NYU Depth V2, Gaussian noise ε=10.0average accuracy: 51.94 -> 55.44Improved
Add uncertainty-aware weighting (ablation)QMF without UAW or regularization · NYU Depth V2, Gaussian noise ε=10.0average accuracy: 51.94 -> 55.14Improved
Add regularization term (ablation)QMF with UAW only · NYU Depth V2, Gaussian noise ε=10.0average accuracy: 55.14 -> 55.81Improved
Add uncertainty-aware weighting (ablation)QMF with regularization only · NYU Depth V2, Gaussian noise ε=10.0average accuracy: 55.44 -> 55.81Improved
Use QMF instead of TMCmultimodal classification · NYU Depth V2, Gaussian noise ε=0.0average accuracy: 71.06 -> 70.09Hurt
Use QMF instead of MMBTmultimodal classification · MVSA, Gaussian noise ε=0.0average accuracy: 78.50 -> 78.07Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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