analiza aiqbench
SITA addresses the intractability of importance sampling in generative modeling by replacing costly divergence computations with fast surrogate likelihoods from energy-based models. This approach enables efficient Boltzmann distribution sampling for larger molecular systems without the computational overhead of conventional simulation methods. The method demonstrates state-of-the-art performance on peptide benchmarks, offering a practical pathway for scaling inference-time annealing in computational chemistry.
Scalable Inference-Time Annealing with Surrogate Likelihood Estimators
arXiv:2605.31498v3 Announce Type: replace Abstract: A long standing challenge in computational chemistry and biophysics is efficiently sampling the Boltzmann distribution of molecules. Advances in generative modeling have been proposed to address the limitations of conventional sampling techniques by eliminating the computational cost of simulation. A promising direction is iteratively finetuning diffusion models along a temperature ladder whereby training data is generated via importance sampling during inference-time annealing. Unfortunately, these methods require computing a divergence over the score field to estimate importance weights, rendering them intractable for larger systems. Here we present scalable inference-time annealing (SITA), which retrains flow-based models to generate samples at progressively lower temperatures using an energy-based model to facilitate fast surrogate likelihoods. We demonstrate state-of-the-art performance on both Alanine Dipeptide and Alanine Tripeptide while avoiding costly divergence terms. Our code is available at https://github.com/countrsignal/sita.git
Abstract:A long standing challenge in computational chemistry and biophysics is efficiently sampling the Boltzmann distribution of molecules. Advances in generative modeling have been proposed to address the limitations of conventional sampling techniques by eliminating the computational cost of simulation. A promising direction is iteratively finetuning diffusion models along a temperature ladder whereby training data is generated via importance sampling during inference-time annealing. Unfortunately, these methods require computing a divergence over the score field to estimate importance weights, rendering them intractable for larger systems. Here we present scalable inference-time annealing (SITA), which retrains flow-based models to generate samples at progressively lower temperatures using an energy-based model to facilitate fast surrogate likelihoods. We demonstrate state-of-the-art performance on both Alanine Dipeptide and Alanine Tripeptide while avoiding costly divergence terms. Our code is available at this https URL
Machine Learning (cs.LG); Biomolecules (q-bio.BM)
arXiv:2605.31498 [cs.LG]
(or arXiv:2605.31498v3 [cs.LG] for this version)
Text integral sub licența deschisă a publicației · 147 cuvinte