Research map / Label-efficient and robust learning / Domain adaptation and generalization
Cross-domain text and graph transfer
Transfers across domains in named entity recognition and graph recommendation via contrastive learning.
54 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2017–2026) · an approach in Domain adaptation and generalization, keyword cluster “cross domain · transfer · domain shot”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
25 papers in 2023–24 and 12 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 7.3% to 3.8%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2017 · Learning to Discover Cross-Domain Relations with Generative Adversarial Networks (ICML 2017 · 726 citations)
- 2018 · Hierarchical Attention Transfer Network for Cross-Domain Sentiment Classification (AAAI 2018 · 167 citations)
- 2019 · Deeply Fusing Reviews and Contents for Cold Start Users in Cross-Domain Recommendation Systems (AAAI 2019 · 167 citations)
- 2020 · Probability Weighted Compact Feature for Domain Adaptive Retrieval (CVPR 2020 · 52 citations)
- 2021 · Embracing Domain Differences in Fake News: Cross-domain Fake News Detection using Multi-modal Data (AAAI 2021 · 152 citations)
- 2022 · Exploring Endogenous Shift for Cross-Domain Detection: A Large-Scale Benchmark and Perturbation Suppression Network (CVPR 2022 · 50 citations)
- 2023 · Unsupervised Cross-Domain Rumor Detection with Contrastive Learning and Cross-Attention (AAAI 2023 · 27 citations)
- 2024 · Graph Disentangled Contrastive Learning with Personalized Transfer for Cross-Domain Recommendation (AAAI 2024 · 32 citations)
- 2025 · DisCo: Graph-Based Disentangled Contrastive Learning for Cold-Start Cross-Domain Recommendation (AAAI 2025 · 18 citations)
- 2026 · From IDs to Semantics: A Generative Framework for Cross-Domain Recommendation with Adaptive Semantic Tokenization (AAAI 2026 · 2 citations)
Earliest papers here
- Learning to Discover Cross-Domain Relations with Generative Adversarial Networks (ICML 2017 · 726 citations)
- Hierarchical Attention Transfer Network for Cross-Domain Sentiment Classification (AAAI 2018 · 167 citations)
- Extracting Relationships by Multi-Domain Matching (NeurIPS 2018 · 62 citations)
What has been tried
46 papers here report 438 measured changes (a component added, removed or swapped, compared with a baseline): helped 82%, no clear effect 2%, hurt 13%, mixed 3%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Add eliminatory contrastive learning | Sport&Phone (Sport) · HR@10 | 54.59 -> 56.73 | Helped | Graph Disentangled Contrastive Learning with Personalized Transfer for Cross-Domain Recommendation (AAAI 2024 · 32 citations) |
| Add Semantic Prompt Units | DomainNet UCDR, Infograph query, Unseen Gallery · mAP@200 | 0.5529 -> 0.5798 | Helped | ProS: Prompting-to-simulate Generalized knowledge for Universal Cross-Domain Retrieval (CVPR 2024 · 23 citations) |
| Add the cross-domain decoder | Sport, Cloth, Game, Video, Music, Movie, Phone, Elec · HR@10 | Sport: 9.81±0.24 -> 10.72±0.32; Cloth: 11.79±0.29 -> 12.85±0.42; Game: 9.20±0.32 -> 9.54±0.28; Video: 11.5±0.31 -> 13.38 | Helped | DisCo: Graph-Based Disentangled Contrastive Learning for Cold-Start Cross-Domain Recommendation (AAAI 2025 · 18 citations) |
| Add cross-domain alignment | DomainNet (average, 12 tasks) · P@200 | 47.02 -> 57.26 | Helped | Unsupervised Cross-Domain Image Retrieval via Prototypical Optimal Transport (AAAI 2024 · 12 citations) |
| Add Lpsu (prototype-score unitization) | OfficeHome, partial rate 0.3 · mAP | 0.477 -> 0.573 | Helped | DiDA: Disambiguated Domain Alignment for Cross-Domain Retrieval with Partial Labels (AAAI 2024 · 11 citations) |
| Add DSFS (domain division) | Aliccp S#1, S#2, and S#3 · AUC | 0.6214 -> 0.6240; 0.5995 -> 0.6000; 0.6175 -> 0.6209 | Helped | D3: A Methodological Exploration of Domain Division, Modeling, and Balance in Multi-Domain Recommendations (AAAI 2024 · 10 citations) |
| replace unified joint training with single-domain learning | MDRAU-MT P1 · R@20 | 0.0545 -> 0.0281 | Hurt | Multi-Domain Recommendation to Attract Users via Domain Preference Modeling (AAAI 2024 · 9 citations) |
| Add feature distillation | Amazon reviews cross-domain classification · accuracy | Helped | TACIT: A Target-Agnostic Feature Disentanglement Framework for Cross-Domain Text Classification (AAAI 2024 · 8 citations) | |
| Add bidirectional state space modeling (BSSM) | Fatigue Fracture (Domain-1→Domain-2) · ACC | 52.7 -> 57.9; +5.2% | Helped | Samba: Severity-aware Recurrent Modeling for Cross-domain Medical Image Grading (NeurIPS 2024 · 5 citations) |
| add CMKD to the backbone-only baseline | ICFG-PEDES → RSTPReid · Rank-1 | 55.00 -> 58.20 | Helped | Graph-Based Cross-Domain Knowledge Distillation for Cross-Dataset Text-to-Image Person Retrieval (AAAI 2025 · 4 citations) |
| Add Gaussian noise to source embeddings | Amazon-16@CDs · HR@5 | 0.5267 -> 0.4963 | Hurt | Federated Graph Learning for Cross-Domain Recommendation (NeurIPS 2024 · 4 citations) |
| Add task-description prompt | CROSSNEWS attribution, Article-Article · accuracy | 52.8 -> 56.9 | Helped | CROSSNEWS: A Cross-Genre Authorship Verification and Attribution Benchmark (AAAI 2025 · 2 citations) |
What reviewers will expect
Datasets and baselines named in the 6 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Baselines
- BPR-MF · 2 of 6 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 37 papers; from OpenAlex and the papers' affiliations).
- Sichuan University · 3 papers
- Alibaba Group · 3 papers
- Delhi Technological University · 2 papers
- A*STAR · 2 papers
- Tsinghua University · 2 papers
- Zhejiang University · 2 papers
Use it
Download BibTeX of 38 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Cross-domain few-shot with transformers (116 papers)
- Mixture and distillation for domain shift (78 papers)
- Statistical transfer learning (79 papers)
Elsewhere in the topic
- Conditional distribution alignment (162 papers)
- Feature alignment for unsupervised adaptation (105 papers)
- Optimization-based domain generalization (101 papers)
- Adversarial domain adaptation (95 papers)
- Geometric alignment and synthetic-to-real adaptation (93 papers)
- Source-free domain adaptation (92 papers)
- Adaptation for semantic segmentation (81 papers)
- Statistical alignment and hypothesis testing (80 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).