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..
Learning Repetition-Invariant Representations for Polymer Informatics
Authors: Yihan Zhu, Gang Liu, Eric Inae, Tengfei Luo, Meng Jiang
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct experiments on four homopolymer and two copolymer datasets. |
| Researcher Affiliation | Academia | Yihan Zhu University of Notre Dame EMAIL Gang Liu University of Notre Dame EMAIL Eric Inae University of Notre Dame EMAIL Tengfei Luo University of Notre Dame EMAIL Meng Jiang University of Notre Dame EMAIL |
| Pseudocode | No | The paper describes the methodology of GRIN through textual descriptions and mathematical equations in Section 3, but does not include any clearly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper mentions 'Code is implemented in PYTHON 3.11 and PYTORCH 2.1.0+cu118, with graph processing using PYG 2.5.1' but does not provide an explicit statement about the public release of their own source code or a link to a repository. |
| Open Datasets | Yes | The first three datasets are extracted from Polymer Info [23], O2Perm is compiled from the Membrane Society of Australasia portal following [29]. ... [23] S. Otsuka, I. Kuwajima, J. Hosoya, Y. Xu, and M. Yamazaki. Polyinfo: Polymer database for polymeric materials design. In 2011 International Conference on Emerging Intelligent Data and Web Technologies, pages 22 29. IEEE, 2011. ... [29] A. W. Thornton, B. D. Freeman, and L. M. Robeson. Polymer gas separation membrane database. https://membrane-australasia.org/ polymer-gas-separation-membrane-database/, 2012. Accessed: 2025-05-11. ... Data are obtained and processed as Aldeghi and Coley [2]. |
| Dataset Splits | Yes | each split into 60% training, 10% validation, and 30% test sets. |
| Hardware Specification | Yes | All experiments were conducted on a 16-core Intel Xeon Gold 6130 CPU (2.1 GHz) with 96 GB RAM and a single NVIDIA A6000 GPU (48 GB). |
| Software Dependencies | Yes | Code is implemented in PYTHON 3.11 and PYTORCH 2.1.0+cu118, with graph processing using PYG 2.5.1. |
| Experiment Setup | Yes | Layer Number {2, 3}, Learning Rate {1e-2, 1e-3}, Batch = 32, Hidden Dimension = 300, ℓ1 Regularization Weight = 1e-3. ... we do not add the ℓ1 sparsity penalty until after 50 epochs during the training process. ... Each model is trained for up to 400 epochs with early stopping (patience=100)... |