Language-Informed Flow Matching for Trend-Guided Structure-Based 3D Molecular Generation
cs.LG
Submitted: 2026-08-31
Updated: 2026-08-31
Comments: Accepted at Findings of EMNLP 2026
Code: https://github.com/kasurl/LiFT
License: http://creativecommons.org/licenses/by/4.0/
The gist: Structure-based drug design (SBDD) requires ligands that satisfy both 3D target affinity and 1D chemical validity.
Terminology
Abstract
Structure-based drug design (SBDD) requires ligands that satisfy both 3D target affinity and 1D chemical validity. Existing controllable generation methods often rely on task-specific fine-tuning or externally imposed sampling-time guidance, adding cost and potentially conflicting with evolving 3D geometric constraints. We propose LiFT, a language-informed cross-modal framework built on Flow Matching for trend-guided 3D molecular generation across both de novo design and scaffold hopping. LiFT uses a "Sense-Evolve-Assemble" agent to generate target-aware SMILES as intermediate chemical conditions, from which a pre-trained chemical foundation model extracts continuous semantic priors. These priors are integrated into geometric generation through a lightweight semantic projector with zero-initialized adaptive normalization for stable cross-modal conditioning. We further introduce a Self-Conditioned Decoupled Router (SCDR), which modulates the velocity field according to intermediate structural states during ODE integration. Experiments on Cross-Docked2020 show that LiFT achieves competitive distribution matching while improving medicinal chemistry metrics and maintaining competitive structural validity under task-steering settings without additional generator fine-tuning. Our results suggest that language-derived chemical priors provide effective trend-level guidance for 3D molecular generation. Code and released artifacts are available at https://github.com/kasurl/LiFT.
Sources
- MolSculpt: Sculpting 3D Molecular Geometries from Chemical Syntax
- Empowering LLMs for Structure-Based Drug Design via Exploration-Augmented Latent Inference
- Chem3DLLM: 3D Multimodal Large Language Models for Chemistry
- Rectified Flow For Structure Based Drug Design
- MolChord: Structure-Sequence Alignment for Protein-Guided Drug Design
- Geometric Deep Learning for Structure-Based Drug Design: A Survey
- Unveiling the Secret of AdaLN-Zero in Diffusion Transformer
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks