Research map / Label-efficient and robust learning / Recognition and long-tail
CONTEXTOR: Contextualized High-order Contrastive Learning
Ze Cai, Hanzhe Liang, Sihang Zeng, Binbin Zhou, Jun Wen
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 ACM (query-conditioned modulation) instead of static projection (ablation) | HGNN+CONTEXTOR · DGA Drug Discovery | MRR: 0.2853 -> 0.3180 | Improved |
| Add ACM instead of static projection (ablation) | HGNN+CONTEXTOR · DGA Gene Identification | MRR: 0.2270 -> 0.2466 | Improved |
| Add ACM instead of static projection (ablation) | HGNN+CONTEXTOR · DGA ADR Prediction | MRR: 0.3845 -> 0.4092 | Improved |
| Add CONTEXTOR loss to MLP backbone | MLP · DGA Drug Discovery | MRR: 0.2481 -> 0.2922 | Improved |
| Add CONTEXTOR loss to GCN backbone | GCN · DGA Drug Discovery | MRR: 0.2655 -> 0.3315 | Improved |
| Add CONTEXTOR loss to HGNN backbone | HGNN · DGA Drug Discovery | MRR: 0.2741 -> 0.3180 | Improved |
| Add CONTEXTOR loss to HGNN backbone | HGNN · DGA Gene Identification | MRR: 0.2144 -> 0.2466 | Improved |
| Add CONTEXTOR loss to MLP backbone | MLP · DGA ADR Prediction | MRR: 0.3296 -> 0.3867 | Improved |
| Replace DGI contrastive loss of MCHNN with CONTEXTOR | MCHNN · DMD Hits@1 (%) | Hits@1: 90.04 -> 93.30 | Improved |
| Replace reconstruction loss of HGSynergy with CONTEXTOR | HGSynergy · NCI-ALMANAC random CV | AUROC: 0.8530 -> 0.8581 | Improved |
| Replace reconstruction loss of HGSynergy with CONTEXTOR | HGSynergy · NCI-ALMANAC cell line level | AUROC: 0.7886 -> 0.7884 | No clear effect |
| Increase contrastive loss weight alpha (ablation) | CONTEXTOR · DGA hyperparameter sweep (alpha 0.0-0.9) | AUROC, AUPRC, MRR | Improved |
Closest papers by meaning
- The Atlas of In-Context Learning: How Attention Heads Shape In-Context Retrieval Augmentation (NeurIPS 2025 · similarity 0.83)
- Separating common from salient patterns with Contrastive Representation Learning (ICLR 2024 · similarity 0.83)
- Synergistic Anchored Contrastive Pre-training for Few-Shot Relation Extraction (AAAI 2024 · similarity 0.83)
- Enhancing Low-Resource Relation Representations through Multi-View Decoupling (AAAI 2024 · similarity 0.83)
- Automatically Generating Numerous Context-Driven SFT Data for LLMs Across Diverse Granularity (AAAI 2025 · similarity 0.83)
- Knowledge Graph Completion with Relation-Aware Anchor Enhancement (AAAI 2025 · similarity 0.82)
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.
