Research map / Label-efficient and robust learning / Domain adaptation and generalization
Rein++: Efficient Generalization and Adaptation for Semantic Segmentation With Vision Foundation Models
Zhixiang Wei, Xiaoxiao Ma, Ruishen Yan, Tao Tu, Huaian Chen, Jinjin Zheng, Yi Jin, Enhong Chen
This paper tried a common change: Larger LoRA rank
raise LoRA rank r from 16 to 4 · GTAV -> Citys.+BDD.+Map. (avg) · mIoU: 67.5 -> 66.8 Hurt
What 233 accepted papers report for this change
- 107 improved
- 60 no clear effect
- 19 mixed
- 47 hurt
- Understanding LoRA as Knowledge Memory: An Empirical Analysis ICML 2026 · increase LoRA rank · CounterFact · efficacyImproved
- SLQ: Bridging Modalities via Shared Latent Queries for Retrieval with Frozen MLLMs ICML 2026 · increase LoRA rank from 4 to 32 · COCO image-to-text · Recall@5: 84.5 -> 82.1Hurt
- Trojan-Speak: Bypassing Constitutional Classifiers with No Jailbreak Tax via Adversarial Finetuning ICML 2026 · Increase LoRA rank (8 to 128) · GPQA-Diamond during Stage 2 · accuracyImproved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
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 |
|---|---|---|---|
| add Rein-G to a frozen DINOv2-Large | Frozen DINOv2-L · GTAV -> Citys.+BDD.+Map. (avg) | mIoU: 65.6 -> 67.5 | Improved |
| fully fine-tune DINOv2-Large instead of freezing | DINOv2-L · GTAV -> Citys.+BDD.+Map. (avg) | mIoU: 65.6 -> 61.7 | Hurt |
| add Rein-G to frozen CLIP-Large | CLIP-L · GTAV -> Citys.+BDD.+Map. (avg) | mIoU: 52.9 -> 59.2 | Improved |
| add Rein-G to frozen DINOv2-Giant | DINOv2-G · GTAV -> Citys.+BDD.+Map. (avg) | mIoU: 65.1 -> 68.4 | Improved |
| add Rein-core (ablation) | Frozen DINOv2-L · GTAV -> Cityscapes | mIoU: 68.0 -> 69.9 | Improved |
| add Rein-link (ablation) | Rein-core · GTAV -> Cityscapes | mIoU: 69.9 -> 70.1 | Improved |
| add multi-head (ablation) | Rein-lora · GTAV -> Cityscapes | mIoU: 70.1 -> 70.8 | Improved |
| add GELU (ablation) | Rein-G with multi-head · GTAV -> Cityscapes | mIoU: 70.8 -> 70.9 | Improved |
| raise LoRA rank r from 16 to 4 (ablation) | Rein-G · GTAV -> Citys.+BDD.+Map. (avg) | mIoU: 67.5 -> 66.8 | Hurt |
| increase token length from 100 to 200 (ablation) | Rein-G · GTAV -> Cityscapes | mIoU: 70.7 -> 68.9 | Hurt |
| add Rein-G to Rein-A (ablation) | Rein-A (DINOv2-L) · GTA -> Cityscapes val | mIoU: 71.5 -> 76.9 | Improved |
| add L_mix loss (ablation) | Rein-A (DINOv2-L) · GTA -> Cityscapes val | mIoU: 71.6 -> 76.9 | Improved |
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.
- FisherTune — FisherTune: Fisher-Guided Robust Tuning of Vision Foundation Models for Domain Generalized Segmentation (CVPR 2025) · better on 1 of 1 shared result
- HRDA — HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentation (ECCV 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
- Stronger Fewer & Superior: Harnessing Vision Foundation Models for Domain Generalized Semantic Segmentation (CVPR 2024 · similarity 0.93)
- Unleashing the Power of Visual Foundation Models for Generalizable Semantic Segmentation (AAAI 2025 · similarity 0.91)
- UniDSeg: Unified Cross-Domain 3D Semantic Segmentation via Visual Foundation Models Prior (NeurIPS 2024 · similarity 0.90)
- Leveraging Depth and Language for Open-Vocabulary Domain-Generalized Semantic Segmentation (NeurIPS 2025 · similarity 0.90)
- Collaborating Foundation Models for Domain Generalized Semantic Segmentation (CVPR 2024 · similarity 0.89)
- Mamba as a Bridge: Where Vision Foundation Models Meet Vision Language Models for Domain-Generalized Semantic Segmentation (CVPR 2025 · similarity 0.88)
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.
