AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

DeepSADR: Deep Transfer Learning with Subsequence Interaction and Adaptive Readout for Cancer Drug Response Prediction

Yuanpeng Zhang, Zhijian Huang, Ziyu Fan, Siyuan Shen, Yahan Li, Shangqian Wu, Min Wu, Lei Deng

ICLR 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors

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 changedWhereMeasuredEffect
add adaptive Set-Transformer readout instead of sum/max/mean (ablation)DeepSADR · TCGA patients (average of 5 drugs)AUC: 0.662 -> 0.856Improved
add adaptive Set-Transformer readout instead of sum/max/mean (ablation)DeepSADR · TCGA patients (average of 5 drugs)AUPR: 0.675 -> 0.862Improved
add subsequence interaction graph module (ablation)DeepSADR · TCGA patients (average of 5 drugs)AUC: 0.698 -> 0.856Improved
add subsequence interaction graph module (ablation)DeepSADR · TCGA patients (average of 5 drugs)AUPR: 0.710 -> 0.862Improved
add threshold selection to prune weak interaction edges (ablation)DeepSADR · TCGA patients (average of 5 drugs)AUC: 0.775 -> 0.856Improved
add threshold selection to prune weak interaction edges (ablation)DeepSADR · TCGA patients (average of 5 drugs)AUPR: 0.749 -> 0.862Improved
feed pre-trained drug response features as extra input to fine-tuning (ablation)DeepSADR · TCGA patients (average of 5 drugs)AUC: 0.781 -> 0.856Improved
feed pre-trained drug response features as extra input to fine-tuning (ablation)DeepSADR · TCGA patients (average of 5 drugs)AUPR: 0.787 -> 0.862Improved
use GNP drug subsequence encoder instead of simplest GNN (ablation)DeepSADR · TCGA patients, SorafenibAUC: 0.907/0.016 -> 0.957/0.037Improved
fine-tune on patient data after pre-training on cell linesDeepSADR · TCGA patients, CisplatinAUC: 0.585 -> 0.927Improved
fine-tune on patient data after pre-training on cell linesDeepSADR · TCGA patients, FluorouracilAUC: 0.534 -> 0.806Improved
use DeepSADR instead of GANDALF on FluorouracilGANDALF · TCGA patients, FluorouracilAUC: 0.793/0.031 -> 0.805/0.056No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most 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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy