TABASCO: A Fast, Simplified Model for Molecular Generation with Improved Physical Quality

2025-07-01

AI Summary

Existing 3D molecular generation models incorporate SE(3)-equivariance and graph-based message passing but suffer from limited physical plausibility and low inference efficiency. This work proposes a minimalist, non-equivariant Transformer architecture that treats atoms as a sequence—omitting both SE(3) constraints and explicit message passing—and reconstructs covalent bonds post-generation via deterministic chemical valence rules. To our knowledge, this is the first purely sequential 3D molecular generator without equivariance or message passing. Remarkably, rotational equivariance emerges implicitly during training, substantially improving geometric validity and chemical feasibility. On GEOM-Drugs, our method achieves state-of-the-art PoseBusters validity while accelerating inference by approximately 10× over the strongest baseline. The approach thus delivers unprecedented efficiency alongside high structural and chemical realism, making it well-suited for structure-based and pharmacophore-driven drug design.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Generation

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

State-of-the-art models for 3D molecular generation are based on significant inductive biases, SE(3), permutation equivariance to respect symmetry and graph message-passing networks to capture local chemistry, yet the generated molecules still struggle with physical plausibility. We introduce TABASCO which relaxes these assumptions: The model has a standard non-equivariant transformer architecture, treats atoms in a molecule as sequences and reconstructs bonds deterministically after generation. The absence of equivariant layers and message passing allows us to significantly simplify the model architecture and scale data throughput. On the GEOM-Drugs benchmark TABASCO achieves state-of-the-art PoseBusters validity and delivers inference roughly 10x faster than the strongest baseline, while exhibiting emergent rotational equivariance despite symmetry not being hard-coded. Our work offers a blueprint for training minimalist, high-throughput generative models suited to specialised tasks such as structure- and pharmacophore-based drug design. We provide a link to our implementation at github.com/carlosinator/tabasco.

Problem

• Improves physical plausibility in 3D molecular generation
• Simplifies model architecture without equivariant layers
• Enables faster inference for drug design tasks

Innovation

• Uses non-equivariant transformer architecture
• Treats atoms as sequences deterministically
• Simplifies model for high throughput