Margin-calibrated Classifier Guidance for Property-driven Synthesis Planning
About
Synthesis planning seeks an efficient sequence of chemical reactions that produce a target molecule. Typically, a pretrained single-step (autoregressive) retrosynthesis model is repeatedly invoked to generate such a sequence. Classifier guidance can, in principle, help steer the output of single-step model toward reactions that satisfy specific constraints or accommodate chemist's preferences during inference without having to retrain the autoregressive generator. We expose the insufficiency of auxiliary classifiers trained with cross-entropy loss to override the unconditional token-level distributions learned from typical sparse single-disconnection reaction datasets. We overcome this issue with a novel method called Sequence Completion Ranking (SCR), which employs contrastive argumentation and a margin-based loss to calibrate the classifier so that it can meaningfully discriminate between continuations during decoding. We formally establish that margin-calibrated classifiers can expand the set of property-satisfying sequences reachable under guided beam search. Empirically, on USPTO-190, given chemist-specified guidance targets, SCR substantially improves multi-step solve rates from $16.8\%$ (unguided generator) to $78.4\%$ with reaction-type guidance and $95.3\%$ with Tanimoto guidance, unlocking valid routes for 33 targets ($17.4\%$) previously unsolvable with baselines. Our method also effectively closes the long-standing diversity gap between template-free and template-based methods.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Multi-step synthesis | USPTO-190 | Success Rate95 | 14 | |
| Multi-step synthesis | Pistachio reachable | Rate97 | 10 | |
| Retrosynthesis Search | Pistachio Hard USPTO-50k (train) | Rate0.89 | 9 | |
| Search-based synthesis planning | USPTO 50k checkpoint 190 (test) | Rate0.95 | 5 | |
| Retrosynthesis Search | Pistachio Hard USPTO-190 context (train) | Success Rate93 | 3 | |
| Search-based synthesis planning | USPTO 190 (test) | Success Rate91 | 3 |