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..

Glocal Information Bottleneck for Time Series Imputation

Authors: Jie Yang, Kexin Zhang, Guibin Zhang, Philip S Yu, Kaize Ding

NeurIPS 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on nine datasets confirm that Glocal-IB leads to consistently improved performance and aligned latent representations under missingness.
Researcher Affiliation Academia 1University of Illinois Chicago 2Northwestern University 3National University of Singapore
Pseudocode No The paper describes methods and derivations using mathematical equations, but does not present any dedicated pseudocode or algorithm blocks.
Open Source Code Yes Our code implementation is available in https://github.com/Muyiiiii/Neur IPS-25-Glocal-IB.
Open Datasets Yes Datasets: Comprehensive experiments are conducted on nine public time-series datasets [59, 75, 26, 73], including ETTh1, ETTh2, ETTm1, ETTm2, Beijing Air, PEMS-Traffic, Electricity, Weather, and Metr-LA.
Dataset Splits Yes We follow the data processing and split protocol from Py POTS [9]. The training, validation, and test sets are divided (60%, 20%, and 20%) in chronological order to avoid data leakage.
Hardware Specification Yes All experiments are implemented in Py Torch [40] 2.6.0 and run on a single NVIDIA 4090 GPU with 24GB memory.
Software Dependencies Yes All experiments are implemented in Py Torch [40] 2.6.0 and run on a single NVIDIA 4090 GPU with 24GB memory. We use the Adam optimizer [22] with a learning rate of 0.001.
Experiment Setup Yes We use the Adam optimizer [22] with a learning rate of 0.001. The batch size is 64, and the number of training epochs is fixed to 30. The hidden dimension is set to 256.