Re-evaluating Retrosynthesis Algorithms with Syntheseus

arXiv:2310.19796v2 Announce Type: replace
Abstract: The planning of how to synthesize molecules, also known as retrosynthesis, has been a growing focus of the machine learning and chemistry communities in recent years. Despite the appearance of steady progress, we argue that imperfect benchmarks and inconsistent comparisons mask systematic shortcomings of existing techniques. To remedy this, we present a benchmarking library called syntheseus which promotes best practice by default, enabling consistent meaningful evaluation of single-step and multi-step retrosynthesis algorithms. We use syntheseus …

abstract algorithms arxiv benchmarks chemistry communities cs.ai cs.lg focus machine machine learning molecules planning progress q-bio.qm type