Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
DAAC: Discrepancy-Aware Adaptive Contrastive Learning for Medical Time series
Authors: Yifan WANG, Hongfeng Ai, ruiqi li, Maowei Jiang, Quangao Liu, Jiahua Dong, ruiyuan kang, Alan Liang, Zihang Wang, ruikai liu, Cheng Jiang, Chenzhong Li
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on four clinical datasets covering Alzheimer s disease, Parkinson s disease, myocardial infarction and alcoholic liver disease demonstrate that DAAC significantly outperforms existing methods, even when only 10% of labeled data is available, showing strong generalization and diagnostic performance. |
| Researcher Affiliation | Academia | Yifan Wang*1, Hongfeng Ai*1, Ruiqi Li*2, Maowei Jiang* 3, Quangao Liu2, Jiahua Dong5, Ruiyuan Kang4, Alan Liang2,6, Zihang Wang2, Ruikai Liu2, Cheng Jiang 1, Chenzhong Li 1 1 School of Medicine, The Chinese University of Hong Kong, Shenzhen, Guangdong, 518172, P.R. China; 2 UCAS; 3 Tsinghua University; 4 TII; 5 MBZUAI; 6 NUS *Equal contribution Project leader Corresponding author |
| Pseudocode | No | The paper describes the proposed methods using textual descriptions and high-level architectural diagrams (e.g., Figure 1), but it does not include any explicit pseudocode blocks or algorithm listings. |
| Open Source Code | Yes | Our code is available at https://github.com/CUHKSZ-MED-BioE/DAAC. |
| Open Datasets | Yes | All datasets are publicly available and had received institutional review board (IRB) approval prior to their release. The AD dataset Miltiadous et al. [2023] is a publicly available EEG time-series dataset ... The PTB-XL dataset Wagner et al. [2020] is a large-scale public ECG time-series dataset... |
| Dataset Splits | Yes | in the subject-dependent setup, 60%, 20%, and 20% of the samples were allocated to the training, validation, and test sets, respectively; in the subject-independent setup, 60%, 20%, and 20% of the subjects (and their corresponding samples) were used for training, validation, and testing, respectively. (from C1.1 Alzheimer s Disease Dataset) |
| Hardware Specification | Yes | All experiments were conducted on a workstation equipped with four NVIDIA RTX 4090 GPUs (24GB each). |
| Software Dependencies | Yes | The models were implemented in PyTorch 1.13.1 with CUDA 11.6. |
| Experiment Setup | Yes | Finally, the overall contrastive loss function L is defined as: L = λS LS + λR LR + λE LE + λT LT + λV LV (12) where λS : λR : λE : λT : λV = 1 : 1 : 1 : 1 : 2 in our practice. The details about loss weight sensitivity could be seen in Appendix D.2 and Table G1: Architecture details of DAAC used in our experiments. |