Research map / Label-efficient and robust learning / Domain adaptation and generalization
Learning List-Level Domain-Invariant Representations for Ranking
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 list-level alignment (ListDA) to zero-shot RankT5 | RankT5 · Robust04 | NDCG@10: 0.5340 -> 0.5573 | Helped |
| Add list-level alignment (ListDA) to zero-shot RankT5 | RankT5 · TREC-COVID | NDCG@10: 0.8200 -> 0.8412 | Helped |
| Add list-level alignment (ListDA) to zero-shot RankT5 | RankT5 · BioASQ | NDCG@10: 0.5542 -> 0.5714 | Helped |
| Replace item-level alignment (ItemDA) with list-level alignment (ListDA) | RankT5 · BioASQ | NDCG@10: 0.5343 -> 0.5714 | Helped |
| Apply item-level alignment (ItemDA) instead of zero-shot | RankT5 · BioASQ | NDCG@10: 0.5542 -> 0.5343 | Hurt |
| Apply item-level alignment (ItemDA) instead of zero-shot | RankT5 · Robust04 | NDCG@10: 0.5340 -> 0.5396 | No clear effect |
| Replace QGen pseudolabels with ListDA using the same synthesized queries | RankT5 · Robust04 | NDCG@10: 0.5034 -> 0.5573 | Helped |
| Replace QGen pseudolabels with ListDA | RankT5 · BioASQ | NDCG@10: 0.5643 -> 0.5714 | No clear effect |
| Replace MLP discriminator with transformer discriminator in ItemDA (ablation) | ItemDA · Robust04, TREC-COVID, BioASQ | all metrics: 0.5396 -> 0.5406 (Robust04 NDCG@10) | No clear effect |
| Use pairwise logistic ranking loss instead of softmax cross-entropy for ListDA (ablation) | ListDA · Robust04 | NDCG@10: 0.5573 -> 0.5524 | Hurt |
| Combine ListDA with QGen PL objective (ablation) | ListDA · Robust04 | NDCG@10: 0.5573 -> 0.5386 | Hurt |
| Combine ListDA with QGen PL objective (ablation) | ListDA · Signal-1M | NDCG@10: 0.2602 -> 0.2817 | Helped |
Closest papers by meaning
- A Domain Generalization Perspective on Listwise Context Modeling (AAAI 2019 · similarity 0.87)
- PiRank: Scalable Learning To Rank via Differentiable Sorting (NeurIPS 2021 · similarity 0.87)
- Adaptive Methods for Real-World Domain Generalization (CVPR 2021 · similarity 0.86)
- Domain Adaptation with Invariant Representation Learning: What Transformations to Learn? (NeurIPS 2021 · similarity 0.86)
- Domain Generalization via Conditional Invariant Representations (AAAI 2018 · similarity 0.85)
- Category-Aware Active Domain Adaptation (ICML 2024 · similarity 0.85)
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.