MolAct: An Agentic RL Framework for Molecular Editing and Property Optimization
This addresses molecular design for drug discovery or materials science by formalizing it as an agentic RL problem, though it appears incremental as it builds on existing RL and tool-use concepts.
The authors tackled molecular editing and optimization by framing them as sequential, tool-guided decisions and introduced MolAct, an agentic reinforcement learning framework with a two-stage training paradigm. MolEditAgent-7B achieved 100, 95, and 98 valid add, delete, and substitute edits in molecular editing, outperforming baselines like DeepSeek-R1, and MolOptAgent-7B surpassed Claude 3.7 on LogP while remaining competitive on solubility in molecular optimization.
Molecular editing and optimization are multi-step problems that require iteratively improving properties while keeping molecules chemically valid and structurally similar. We frame both tasks as sequential, tool-guided decisions and introduce MolAct, an agentic reinforcement learning framework that employs a two-stage training paradigm: first building editing capability, then optimizing properties while reusing the learned editing behaviors. To the best of our knowledge, this is the first work to formalize molecular design as an Agentic Reinforcement Learning problem, where an LLM agent learns to interleave reasoning, tool-use, and molecular optimization. The framework enables agents to interact in multiple turns, invoking chemical tools for validity checking, property assessment, and similarity control, and leverages their feedback to refine subsequent edits. We instantiate the MolAct framework to train two model families: MolEditAgent for molecular editing tasks and MolOptAgent for molecular optimization tasks. In molecular editing, MolEditAgent-7B delivers 100, 95, and 98 valid add, delete, and substitute edits, outperforming strong closed "thinking" baselines such as DeepSeek-R1; MolEditAgent-3B approaches the performance of much larger open "thinking" models like Qwen3-32B-think. In molecular optimization, MolOptAgent-7B (trained on MolEditAgent-7B) surpasses the best closed "thinking" baseline (e.g., Claude 3.7) on LogP and remains competitive on solubility, while maintaining balanced performance across other objectives. These results highlight that treating molecular design as a multi-step, tool-augmented process is key to reliable and interpretable improvements.