AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

Cross-Sample Augmented Test-Time Adaptation for Personalized Intraoperative Hypotension Prediction

Kanxue Li, Yibing Zhan, Hua Jin, Chongchong Qi, Xu Lin, Baosheng Yu

AAAI 2026 · Label-efficient and robust learning › Test-time adaptation · approach: Pseudo-label test-time adaptation · 1 citations (OpenAlex)

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 CSA-TTA test-time adaptationUniTS · VitalDB (30S, zero-shot)Recall: 43.24 -> 50.70Improved
add CSA-TTA test-time adaptationUniTS · In-hospital (zero-shot)Recall: 43.77% → 53.33%Improved
add CSA-TTA test-time adaptationTimesFM · VitalDB (30S, zero-shot)MSE: 92.77 → 85.28Improved
add CSA-TTA test-time adaptationUniTS · VitalDB (30S, zero-shot)Precision: 71.14 -> 66.56Hurt
add CSA-TTA instead of TTT++TimesFM · VitalDB (2S, fine-tuned)F1: 64.10 -> 64.83Improved
add TTT/TTT++ baseline adaptationTimesFM · VitalDB (2S, fine-tuned)F1: 64.20 -> 64.00No clear effect
add supervised prediction loss to reconstruction-only adaptation (ablation)TimesFM · VitalDB (10min, fine-tuned)F1: 64.60 -> 64.70Improved
add masked reconstruction loss to prediction-only adaptation (ablation)TimesFM · VitalDB (10min, fine-tuned)MSE: 80.49 -> 77.60Improved
add multi-task objective to Recon-only (ablation)TimesFM · VitalDB (5min, fine-tuned)Recall: 71.90 -> 71.60Hurt
increase retrieval Top-K from 1 to 3 (ablation)TimesFM · VitalDB (zero-shot)MSE: 86.96 to 85.28Improved
add cross-sample bank (ablation)TimesFM · VitalDB (zero-shot)F1: 64.47% → 64.97%Improved
add perturbation augmentation to bank (ablation)TimesFM · VitalDB (fine-tuned)MSE: 74.19 → 72.93Improved

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