Coverage Before Control: Route-Instruction Grounding and Steering for Controllable Retrosynthesis

arXiv:2609.39955 · cs.AI, cs.LG · Submitted 2026-09-30 · Read on arXiv

cs.AI, cs.LG

Submitted: 2026-09-30

Updated: 2026-09-30

Code: https://github.com/LOGO-CUHKSZ/RIGS

Terminology

• Grounding — 1× in this paper · explained in 3 episodes
• Steering — 1× in this paper · explained in 2 episodes

Sources

• Learning to Make Generalizable and Diverse Predictions for Retrosynthesis
• From Feasible to Practical: Pareto-Optimal Synthesis Planning
• T-Rex: Text-assisted Retrosynthesis Prediction
• Project Ariadne: Prompt-Conditioned Route Generation for Synthesis Planning
• Synthelite: Chemist-aligned and feasibility-aware synthesis planning with LLMs
• When Single Answer Is Not Enough: Rethinking Single-Step Retrosynthesis Benchmarks for LLMs
• Retrosynthesis prediction enhanced by in-silico reaction data augmentation
• Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• Agents' Last Exam
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems