Reformulates molecular discovery as an autonomous MCTS planning problem over executable chemical operations rather than just similarity-based prediction.
March 26, 2026
Original Paper
MolEvolve: LLM-Guided Evolutionary Search for Interpretable Molecular Optimization
arXiv · 2603.24382
The Takeaway
MolEvolve addresses 'activity cliffs' in chemistry—where small structural changes cause massive property shifts—by using an LLM to evolve symbolic operations. It replaces black-box embeddings with transparent reasoning chains that chemists can actually interpret and execute.
From the abstract
Despite deep learning's success in chemistry, its impact is hindered by a lack of interpretability and an inability to resolve activity cliffs, where minor structural nuances trigger drastic property shifts. Current representation learning, bound by the similarity principle, often fails to capture these structural-activity discontinuities. To address this, we introduce MolEvolve, an evolutionary framework that reformulates molecular discovery as an autonomous, look-ahead planning problem. Unlike