Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
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.
SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks
Authors: Fabian Fuchs, Daniel Worrall, Volker Fischer, Max Welling
NeurIPS 2020 | Venue PDF | LLM Run Details | Input Tokens: 17,501 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 5,254 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate our model on a toy N-body particle simulation dataset, showcasing the robustness of the predictions under rotations of the input. We further achieve competitive performance on two real-world datasets, Scan Object NN and QM9. |
| Researcher Affiliation | Collaboration | Fabian B. Fuchs Bosch Center for Artificial Intelligence A2I Lab, Oxford University... Daniel E. Worrall Amsterdam Machine Learning Lab, Philips Lab University of Amsterdam... Volker Fischer Bosch Center for Artificial Intelligence... Max Welling Amsterdam Machine Learning Lab University of Amsterdam |
| Pseudocode | No | The paper describes the steps of the SE(3)-Transformer in Section 3 and visualizes them in Figure 2, but it does not include a formal pseudocode block or algorithm listing. |
| Open Source Code | Yes | Code available at https://github.com/Fabian Fuchs ML/se3-transformer-public |
| Open Datasets | Yes | We evaluate our model on a toy N-body particle simulation dataset, showcasing the robustness of the predictions under rotations of the input. We further achieve competitive performance on two real-world datasets, Scan Object NN and QM9. ... N-body problem ... adaptation of the dataset from Kipf et al. [14]. ... Scan Object NN ... a recently introduced dataset for real-world object classification. ... The QM9 regression dataset [21] is a publicly available chemical property prediction task. |
| Dataset Splits | No | The paper frequently mentions using 'training' and 'test' sets (e.g., 'training with data augmentation', 'test set'). However, it does not explicitly state details about training/validation/test dataset splits, such as specific percentages, sample counts, or cross-validation methods for all datasets. |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., GPU models, CPU models, memory) used to run the experiments. |
| Software Dependencies | No | The paper mentions a 'Pytorch implementation of spherical harmonics' but does not specify version numbers for PyTorch or any other software dependencies, making the setup not fully reproducible. |
| Experiment Setup | No | The paper describes some architectural choices for the models used in experiments, such as '4 equivariant layers, each followed by an attentive self-interaction layer' or '4 equivariant layers with linear self-interaction followed by max-pooling and an MLP'. However, it does not explicitly provide specific hyperparameter values (e.g., learning rate, batch size, optimizer settings) or system-level training settings in the main text, often deferring 'details' to the appendix. |