Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note3 min read

Autoregressive Boltzmann Generators

Source

arXiv

arxiv.org/abs/2606.27361v1

Read the source

Source

Plain-Language Brief

A new method called Autoregressive Boltzmann Generators (ArBG) is proposed to efficiently sample molecular systems at thermodynamic equilibrium. This method overcomes limitations of previous methods by using an autoregressive modeling framework, which allows for sequential inference-time interventions and leverages architectures effective in Large Language Models. The new method shows significant improvements over flow-based models across all benchmarks, particularly in larger peptide systems like the 10-residue Chignolin. A large transferable model named Robin, trained with ArBG framework, also improves upon the previous state-of-the-art by reducing the zero-shot energy error on 8-residue systems by over 60%. The code for this method can be found at the provided link.

What the Document Says

  • A new method called Autoregressive Boltzmann Generators (ArBG) is proposed.
  • The new method offers enhanced scalability and enables sequential inference-time interventions.
  • ArBG leverages architectures effective in Large Language Models.
  • ArBG leads to significant improvements over flow-based models across all benchmarks, particularly in larger peptide systems like the 10-residue Chignolin.

Key Concepts Mentioned

  • Autoregressive Boltzmann Generators (ArBG)
  • Flow-based models
  • Large Language Models
  • Molecular systems
  • Thermodynamic equilibrium

Claims Made by the Document

  • ArBG offers enhanced scalability and enables sequential inference-time interventions.
  • ArBG leads to significant improvements over flow-based models.

Evidence Used

  • The paper describes the proposed method and its benefits.

Methods / Approach

  • Autoregressive modelling framework
  • Large Language Models

Limitations Stated by the Authors

  • Not stated in source

Limitations Noted by the Agent

Agent interpretation — not from the source.

  • Unknown if the method has any limitations other than those mentioned by the authors.
  • Potential limitations may include the applicability of the method to systems outside of molecular systems at thermodynamic equilibrium.

Open Questions

  • What are the specific architectures used in Large Language Models that ArBG leverages?
  • How does ArBG perform in systems outside of molecular systems at thermodynamic equilibrium?

Notes for Future Q&A

May include interpretation. Labeled where it does.

  • The paper does not provide detailed information about the specific architectures used in Large Language Models.
  • The paper focuses on molecular systems at thermodynamic equilibrium and does not discuss its applicability to other systems.

Raw Abstract / Excerpt

Efficient sampling of molecular systems at thermodynamic equilibrium is a hallmark challenge in statistical physics. This challenge has driven the development of Boltzmann Generators (BGs), which allow rapid generation of uncorrelated equilibrium samples by combining a generative model with exact likelihoods and an importance sampling correction. However, modern BGs predominantly rely on normalizing flows (NFs), which either suffer from limited expressivity due to strict invertibility constraints (discrete time) or computationally expensive likelihoods (continuous time). In this paper, we propose Autoregressive Boltzmann Generators (ArBG) -- a novel autoregressive modelling framework -- that overcomes these limitations by departing from the flow-based BG paradigm. ArBG circumvents the topological constraints of flows and enables sequential inference-time interventions, while offering enhanced scalability by leveraging architectures effective in Large Language Models. We empirically demonstrate that ArBG leads to significant improvements over flow-based models across all benchmarks, but particularly in larger peptide systems such as the 10-residue Chignolin. Furthermore, we introduce Robin, a 132 million parameter transferable model trained with the ArBG framework which improves over the previous state-of-the-art, reducing the zero-shot energy error, E-W$_2$, on 8-residue systems by over 60$%$. The code can be found at the following link: https://github.com/danyalrehman/autobg.