Research map / Label-efficient and robust learning / Domain adaptation and generalization
Unified Modality Separation: A Vision-Language Framework for Unsupervised Domain Adaptation
Xinyao Li, Jingjing Li, Zhekai Du, Lei Zhu, Heng Tao Shen
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 MDI active selection instead of energy (ablation) | UniMoS++ ADA · Office-Home | accuracy: 80.3 -> 81.2 | Improved |
| use MaE ensemble weights in full design vs fixed w*=0.3 (ablation) | UniMoS++ ADA · Office-Home | accuracy: 81.9 -> 82.4 | Improved |
| add L_ortho (ablation) | UniMoS++ ADA · Office-Home | accuracy: 80.7 -> 80.9 | Improved |
| add L_im (ablation) | UniMoS++ ADA · Office-Home | accuracy: 80.9 -> 81.6 | Improved |
| add L_bce (ablation) | UniMoS++ ADA · Office-Home | accuracy: 81.6 -> 81.7 | Improved |
| add learnable w (ablation) | UniMoS++ ADA · Office-Home | accuracy: 81.7 -> 82.0 | Improved |
| add KD to training (ablation) | UniMoS++ ADA · Office-Home | accuracy: 82.0 -> 82.4 | Improved |
| random active selection instead of MDI (ablation) | UniMoS++ ADA · Office-Home | accuracy: 82.4 -> 79.7 | Hurt |
| cross-modality attention instead of MaE (ablation) | UniMoS++ UDA · Office-Home | accuracy: 78.0 -> 77.2 | Hurt |
| max-entropy RL modules instead of MaE (ablation) | UniMoS++ UDA · Office-Home | accuracy: 78.0 -> 76.8 | Hurt |
| add CoOp prompt tuning (ablation) | UniMoS++ ADA · Office-Home 5% | accuracy: 82.4 -> 82.3 | No clear effect |
| add LoRA fine-tuning (ablation) | UniMoS++ ADA · Mini-DomainNet 5% | accuracy: 82.4 -> 82.4 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- MHPL — MHPL: Minimum Happy Points Learning for Active Source Free Domain Adaptation (CVPR 2023) · better on 1 of 1 shared result
- Meta-DMoE — Meta-DMoE: Adapting to Domain Shift by Meta-Distillation from Mixture-of-Experts (NeurIPS 2022) · 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
- Split to Merge: Unifying Separated Modalities for Unsupervised Domain Adaptation (CVPR 2024 · similarity 0.96)
- Breaking Modality Heterogeneity in Low-Bit Quantization for Large Vision-Language Models (NeurIPS 2026 · similarity 0.91)
- Post-pre-training for Modality Alignment in Vision-Language Foundation Models (CVPR 2025 · similarity 0.90)
- Strong Helps Weak: Directional Cross-Modal Alignment Transfer in Multi-modal LLMs (NeurIPS 2026 · similarity 0.90)
- LoMo: Local Modality Substitution for Deeper Vision-Language Fusion (NeurIPS 2026 · similarity 0.89)
- Modality-Specialized Synergizers for Interleaved Vision-Language Generalists (ICLR 2025 · similarity 0.89)
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.
