Learning-Order Autoregressive Models with Application to Molecular Graph Generation
Authors: Zhe Wang, Jiaxin Shi, Nicolas Heess, Arthur Gretton, Michalis Titsias
ICML 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate experimentally that our method can learn meaningful autoregressive orderings in image and graph generation. On the challenging domain of molecular graph generation, we achieve state-of-the-art results on the QM9 and ZINC250k benchmarks, evaluated across key metrics for distribution similarity and drug-likeless. |
| Researcher Affiliation | Collaboration | 1Google Deep Mind 2University College London. Correspondence to: Zhe Wang <EMAIL>, Jiaxin Shi <EMAIL>, Michalis K. Titsias <EMAIL>. |
| Pseudocode | Yes | Algorithm 1 Training with LO-ARM Algorithm 2 Unconditional sampling from LO-ARM |
| Open Source Code | No | The paper does not contain any explicit statement about releasing source code for the described methodology, nor does it provide a link to a code repository. |
| Open Datasets | Yes | On the challenging domain of molecular graph generation, we achieve state-of-the-art results on the QM9 and ZINC250k benchmarks, evaluated across key metrics for distribution similarity and drug-likeless. |
| Dataset Splits | Yes | We follow the standard setup e.g., in Eichelsbacher & Reinert (2008); Vignac et al. (2023); Jo et al. (2022), including data preprocessing, network parametrization, and evaluation metrics. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments. |
| Software Dependencies | No | The paper mentions 'Optimizer Adam W' in Table 7, which is an optimizer, but does not provide specific software names with version numbers for libraries or frameworks used (e.g., Python, PyTorch, TensorFlow versions). |
| Experiment Setup | Yes | Table 7. Hyperparameter setup. Hyperparameter QM9 ZINC250k Optimizer Adam W Adam W Scheduler Cosine Annealing Cosine Annealing Learning Rate 1 10 5 1.5 10 5 Weight Decay 1 1012 1 1012 EMA 0.9999 0.9999 |