Dear authors,
Thank you for sharing your excellent work, "Unified Generative Modeling of 3D Molecules via Bayesian Flow Networks", published at ICLR 2024. Your approach, GeoBFN, demonstrates impressive results in 3D molecular generation tasks.
I would like to inquire if you could provide pre-trained checkpoints for GeoBFN, especially for the experiments on the QM9 and GEOM-DRUG datasets. Having access to these would be extremely beneficial for reproducing your results and conducting further research based on your framework.
If the checkpoints are not yet available, do you have plans to release them in the near future?
Looking forward to your response!
Best regards,
Yuqi
Dear authors,
Thank you for sharing your excellent work, "Unified Generative Modeling of 3D Molecules via Bayesian Flow Networks", published at ICLR 2024. Your approach, GeoBFN, demonstrates impressive results in 3D molecular generation tasks.
I would like to inquire if you could provide pre-trained checkpoints for GeoBFN, especially for the experiments on the QM9 and GEOM-DRUG datasets. Having access to these would be extremely beneficial for reproducing your results and conducting further research based on your framework.
If the checkpoints are not yet available, do you have plans to release them in the near future?
Looking forward to your response!
Best regards,
Yuqi