Research map / Label-efficient and robust learning / Domain adaptation and generalization
TITAN: Query-Token based Domain Adaptive Adversarial Learning
Tajamul Ashraf, Janibul Bashir
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 |
|---|---|---|---|
| add query-based adversarial alignment on encoder and decoder (ablation) | FocalNet-DINO student-teacher baseline · C2F | mAP: 35.2 -> 40.8 | Improved |
| add token-wise adversarial alignment on encoder and decoder (ablation) | FocalNet-DINO student-teacher baseline · C2F | mAP: 35.2 -> 39.7 | Improved |
| combine query-based and token-wise adversarial learning on the encoder (ablation) | FocalNet-DINO · C2F | mAP: 38.5 -> 47.1 | Improved |
| combine query-based and token-wise adversarial learning on the decoder (ablation) | FocalNet-DINO · C2F | mAP: 37.9 -> 45.6 | Improved |
| add variance-based target division into easy and hard subsets (ablation) | TITAN with all adversarial modules · C2F | mAP: 47.1 -> 51.2 | Improved |
| apply TITAN versus best prior source-free method | TITAN · C2F | mAP: 40.9 -> 52.2 | Improved |
| apply TITAN versus UDA method PT on K2C | TITAN · K2C | Car AP: 55.1 -> 53.2 | No clear effect |
| apply TITAN on mammography adaptation versus best source-free baseline | TITAN · RSNA-BSD1K to INBreast | Recall at 0.3 FPI: 0.25 -> 0.78 | Improved |
| use simple three-layer MLP discriminators | TITAN adversarial modules · generalization analysis | generalization bound | Improved |
Later papers that compared with this method
2 later accepted papers report TITAN as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- MOOZY: A Patient-First Foundation Model for Computational Pathology (ECCV 2026 · better on 1 of 5 shared results)
- TopoSlide: Topologically-Informed Histopathology Whole Slide Image Representation Learning (CVPR 2026 · better on 3 of 4 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
- Cross-Domain Adaptive Teacher for Object Detection (CVPR 2022 · similarity 0.91)
- Enhancing Source-Free Domain Adaptive Object Detection with Low-confidence Pseudo Label Distillation (ECCV 2024 · similarity 0.90)
- Large Self-Supervised Models Bridge the Gap in Domain Adaptive Object Detection (CVPR 2025 · similarity 0.89)
- Domain Adaptive Object Detection via Balancing Between Self-Training and Adversarial Learning (TPAMI 2023 · similarity 0.89)
- Self-Training and Adversarial Background Regularization for Unsupervised Domain Adaptive One-Stage Object Detection (ICCV 2019 · similarity 0.89)
- SSAL: Synergizing between Self-Training and Adversarial Learning for Domain Adaptive Object Detection (NeurIPS 2021 · 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.
