mlsanityData to 14 Oct 2026

Continuous Variational Synthesis

arXiv 2609.35083 ↗First posted 28 Sep 2026q-bio.BM cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Biological machine learning was long bottlenecked by the ability to synthesize designed DNA. Variational synthesis models control chemical reactions to physically manufacture quadrillions of designed sequences in DNA. However, training these generative models is challenging: constraints on chemical synthesis can force many parameters into a discrete space, limiting the ability to pre-train and fine-tune. In this article we train ``free'' variational synthesis models using stochastic gradient descent in continuous space, and then discretize with post-training quantization to impose hardware and wetware constraints. This enables variational synthesis models to satisfy stringent reward criteria, while still synthesizing diverse designs, achieving a strictly dominating quality-diversity Pareto frontier. We demonstrate by training variational synthesis models of enzymes, peptides, antibody CDRH3s, and regulatory DNA elements. In silico performance is maintained in vitro.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 18 of its 53 references that are in the map. First posted on arXiv 28 Sep 2026.

Builds on

18 of 53 references.

References · 18

  1. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,898
  2. 2022SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Modelscited by 1,386
  3. 2018Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Reviewcited by 482
  4. 2021A White Paper on Neural Network Quantizationcited by 472
  5. 2016A Kernelized Stein Discrepancy for Goodness-of-fit Tests and Model Evaluationcited by 236
  6. 2023Practical and Asymptotically Exact Conditional Sampling in Diffusion Modelscited by 232
  7. 2022RL with KL penalties is better viewed as Bayesian inferencecited by 135
  8. 2017Measuring Sample Quality with Kernelscited by 119
  9. 2024Probabilistic Inference in Language Models via Twisted Sequential Monte Carlocited by 90
  10. 2022Variational inference via Wasserstein gradient flowscited by 85
  11. 2019Fast and Simple Natural-Gradient Variational Inference with Mixture of Exponential-family Approximationscited by 51
  12. 2023A Rigorous Link between Deep Ensembles and (Variational) Bayesian Methodscited by 32
6 more
  1. 2020Handling the Positive-Definite Constraint in the Bayesian Learning Rulecited by 31
  2. 2022A Unified Perspective on Natural Gradient Variational Inference with Gaussian Mixture Modelscited by 22
  3. 2023Biological Sequence Kernels with Guaranteed Flexibilitycited by 7
  4. 2025Variational Inference with Mixtures of Isotropic Gaussianscited by 6
  5. 2025Accelerated Learning on Large Scale Screens using Generative Library Modelscited by 2
  6. 2025Lifting Biomolecular Data Acquisitioncited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025A Variational Perspective on Generative Protein Fitness Optimization
  2. 2026Constrained Flow Optimization via Sequential Fine Tuning for Molecular Design
  3. 2026On the Design Space of Discrete Diffusion Online Adaptation for Molecular Optimization
  4. 2025Reward-Guided Iterative Refinement in Diffusion Models at Test-Time with Applications to Protein and DNA Design
  5. 2026PairIT: Autoregressive Transformers for Low-Data Molecule Optimization
  6. 2026Training Large Language Models for Small-Molecule Design with Synthetic Task Scaling
  7. 2024Human-level molecular optimization driven by mol-gene evolution
  8. 2024It Takes Two to Tango: Directly Optimizing for Constrained Synthesizability in Generative Molecular Design