Research map / Label-efficient and robust learning / Domain adaptation and generalization
Boomda: Balanced Multi-objective Optimization for Multimodal Domain Adaptation
Jun Sun, Xinxin Zhang, Simin Hong, Jian Zhu, Xiang Gao
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use Boomda instead of CDAN | multimodal domain adaptation · IEMOCAP avg | weighted F1: 46.95 -> 48.73 | Improved |
| use Boomda instead of CDAN | multimodal domain adaptation · MSP-IMPROV avg | weighted F1: 40.07 -> 41.50 | Improved |
| use Boomda instead of PCL on all-modality input | multimodal domain adaptation · IEMOCAP AVL | weighted F1: 49.72 -> 54.82 | Improved |
| Boomda versus other methods on acoustic-visual | multimodal domain adaptation · MSP-IMPROV AV | weighted F1: DANN 42.44 vs Boomda 40.31 | Hurt |
| add correlation alignment (ablation) | Boomda · IEMOCAP avg | weighted F1: 36.75 -> 45.55 | Improved |
| add pseudo labeling by voting (ablation) | Boomda · IEMOCAP avg | weighted F1: 36.75 -> 45.09 | Improved |
| combine correlation alignment and pseudo labeling (ablation) | Boomda · IEMOCAP avg | weighted F1: 45.55 -> 48.73 | Improved |
| use MGDA-based modality-balanced alignment weights instead of equal weights (ablation) | Boomda · IEMOCAP avg | weighted F1: 46.27 -> 48.73 | Improved |
| use balanced alignment weights instead of equal weights (ablation) | Boomda · IEMOCAP VL | weighted F1: 39.91 -> 42.83 | Improved |
| approximate Q with its diagonal matrix for closed-form weights (ablation) | Boomda · IEMOCAP training | ratio r of off-diagonal to diagonal | No clear effect |
| add more modalities to baseline DALN without balancing | DALN · IEMOCAP | weighted F1: 47.23 -> 45.58 | Hurt |
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.
- DADA — DADA: Depth-Aware Domain Adaptation in Semantic Segmentation (ICCV 2019) · better on 1 of 1 shared result
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
- Source-Free Domain Adaptation with Frozen Multimodal Foundation Model (CVPR 2024 · similarity 0.87)
- Rethinking Multimodal Learning from the Perspective of Mitigating Classification Ability Disproportion (NeurIPS 2025 · similarity 0.87)
- Multimodal Representation Learning by Alternating Unimodal Adaptation (CVPR 2024 · similarity 0.87)
- MODfinity: Unsupervised Domain Adaptation with Multimodal Information Flow Intertwining (CVPR 2025 · similarity 0.87)
- Asymmetric Reinforcing Against Multi-Modal Representation Bias (AAAI 2025 · similarity 0.87)
- MMPareto: Boosting Multimodal Learning with Innocent Unimodal Assistance (ICML 2024 · similarity 0.86)
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.
