Skip to content
Conference Open access

AgentRetro: Agent-Enhanced Molecular Spectral Domain Generalization Framework for Retrosynthesis Prediction Under Mixed OOD Shifts

Sep 2026 · Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence · pp. 340-347 · 0 citations · 20 references

TL;DR

AgentRetro is a framework integrating spectral domain generalization with Large Language Model (LLM) agents that consistently improves state-of-the-art baselines, and reveals that explicitly decoupling structural invariance from mechanistic adaptability is essential for overcoming the interference between conflicting distribution shifts.

Abstract

Retrosynthesis prediction is a cornerstone of drug discovery, enabling the synthesis of novel therapeutic candidates. However, current deep learning models falter when navigating the unexplored chemical space essential for innovation. In these realistic scenarios, models face a challenging mixed out-of-distribution (OOD) shift: they must handle both novel molecular scaffolds (covariate shift) and unseen reaction templates (label shift), leading to severe performance degradation. To address this, we propose AgentRetro, a framework integrating spectral domain generalization with Large Language Model (LLM) agents. It features a Spectral Domain Enhancer (SDE) that partitions chemical space via spectral clustering to capture diverse domain structures, and a Multi-Knowledge Agent (MKA) that enriches data with LLM-generated OOD reactions and textual rationales. These components are unified by a dual training strategy: meta-learning for template adaptability and adversarial learning for structural invariance. Crucially, our analysis reveals that explicitly decoupling structural invariance from mechanistic adaptability is essential for overcoming the interference between conflicting distribution shifts. Experiments show AgentRetro consistently improves state-of-the-art baselines, notably boosting EditRetro’s OOD Top-10 accuracy from 41.7% to 50.1%. The source code is available at https://github.com/jiahangshen/IJCAI26-AgentRetro.

Read PDF

Similar papers

Open access Sep 2026

DIAM: Adaptive Drug Repositioning with Decoupled Biological Mechanism and Instance-Aware Modulation

DIAM is proposed, which establishes a mechanism-adaptive paradigm by explicitly decoupling structural and molecular signals by introducing a Dual-Stream Biological Mechanism Decoupling module to construct global structural propagation and local molecular interaction views explicitly.

Ke-Yuan Xu, Ke-Yuan Xu, Shu-Hui Yin et al. · 0 citations
Sep 2026

CRAFT: A Chemical Reaction-Aware Transformer for Synthesizable Molecular Optimization via Multimodal Learning.

CRAFT is a reaction-aware molecular optimization framework that integrates Monte Carlo Tree Search (MCTS) with a multimodal strategy network and explicitly encodes two-dimensional molecular topology and atom-level reaction roles, enabling chemically feasible transformations during multi-step optimization.

Run-Fu Yu, Tong-Mao Ma, Zian Song et al. · 0 citations
Oct 2026

Interpretable Protein-Nucleic Acid Binding Site Prediction via Domain-Guided Multimodal Learning.

BiMTBind is a domain-guided multimodal deep learning framework that integrates bidirectional state-space modeling with Transformer-based attention for residue-level binding site prediction, offering a scalable and generalizable approach for protein-nucleic acid binding site prediction.

Wen-Yu Xi, Ru-Heng Wang, Xiu-Cai Ye et al. · 0 citations
Sep 2026

MLI-DTA: An interpretable multimodal framework with multi-level interactions for drug-target affinity prediction.

MLI-DTA is proposed, a hierarchical multi-source feature fusion network based on multi-level attention that encodes both sequence and graph modalities of drugs and targets, and performs exceptionally well in the Davis and KIBA benchmarks.

Yu-Ning Liu, Guang-Ze Wang, Dan Liu et al. · 0 citations
Review Sep 2026

Artificial Intelligence-Based Halogenated Chalcone Drug Discovery: An Integrated Framework for Rational Anticancer Lead Optimization

Halogenated chalcones are emerging as an exciting class of potential anticancer scaffolds with respect to medicinal chemistry; they exhibit structural flexibility, multitarget pharmacology, and allow for rapid medicinal chemistry optimization. The biological activity associated with halogen substitutions on the scaffol...

Tamara Bassam Kamal Hijjawi · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.