Research map / Label-efficient and robust learning / Recognition and long-tail
Functionally Regionalized Knowledge Transfer for Low-resource Drug Discovery
Huaxiu Yao, Ying Wei, Long-Kai Huang, Ding Xue, Junzhou Huang, Zhenhui Li
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 FRML functional-region localization on ANIL++ | ANIL++ · ChEMBL Assay Group I | mean R2: 0.367 -> 0.375 | Improved |
| add FRML on ANIL++ | ANIL++ · ChEMBL Assay Group II | median R2: 0.252 -> 0.311 | Improved |
| add FRML on ANIL | ANIL · ChEMBL Assay Group I | mean R2: 0.299 -> 0.310 | Improved |
| remove contrastive loss (ablation) | FRML-ANIL++ · ChEMBL Assay Group I | median R2: 0.315 -> 0.328 | Improved |
| remove contrastive loss (ablation) | FRML-ANIL++ · ChEMBL Assay Group I | mean R2: 0.371 -> 0.375 | Improved |
| remove region localization (same capacity) (ablation) | FRML-ANIL++ · ChEMBL Assay Group I | median R2: 0.301 -> 0.328 | Improved |
| replace recurrent localization network with fully connected layers (ablation) | FRML-ANIL++ · ChEMBL Assay Group II | median R2: 0.299 -> 0.311 | Improved |
| increase number of functional regions beyond a few | FRML-ANIL++ · ChEMBL Assay Groups I-IV | mean R2 | No clear effect |
| use only one functional region | FRML-ANIL++ · ChEMBL Assay Groups I-IV | mean R2 | Hurt |
| down-sample support set to 5%, 25%, 50% | FRML-ANIL++ vs ANIL++ · ChEMBL Assay Group I | mean R2 | Improved |
| add FRML on ANIL++ | ANIL++ · MUV | accuracy (%): 56.95 -> 60.66 | Improved |
| add FRML on ANIL++ | ANIL++ · SIDER | accuracy (%): 68.04 -> 70.01 | Improved |
Closest papers by meaning
- Drug-TTA: Test-Time Adaptation for Drug Virtual Screening via Multi-task Meta-Auxiliary Learning (ICML 2025 · similarity 0.85)
- Efficient Biological Data Acquisition through Inference Set Design (ICLR 2025 · similarity 0.84)
- Monroe: A Molecular Foundation Model for In-context Probabilistic Inference (NeurIPS 2026 · similarity 0.83)
- Rethinking the generalization of drug target affinity prediction algorithms via similarity aware evaluation (ICLR 2025 · similarity 0.83)
- AANet: Virtual Screening under Structural Uncertainty via Alignment and Aggregation (NeurIPS 2025 · similarity 0.83)
- MF-LAL: Drug Compound Generation Using Multi-Fidelity Latent Space Active Learning (ICML 2025 · similarity 0.83)
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.
