{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/energy-based-view-of-retrosynthesis","title":"Energy-based View of Retrosynthesis","arxiv_id":"2007.13437","date":"2020-07-14","proceeding":null,"authors":["Ruoxi Sun","Hanjun Dai","Li Li","Steven Kearnes","Bo Dai"],"abstract":"Retrosynthesis -- the process of identifying a set of reactants to synthesize a target molecule -- is of vital importance to material design and drug discovery. Existing machine learning approaches based on language models and graph neural networks have achieved encouraging results. In this paper, we propose a framework that unifies sequence- and graph-based methods as energy-based models (EBMs) with different energy functions. This unified perspective provides critical insights about EBM variants through a comprehensive assessment of performance. Additionally, we present a novel dual variant within the framework that performs consistent training over Bayesian forward- and backward-prediction by constraining the agreement between the two directions. This model improves state-of-the-art performance by 9.6% for template-free approaches where the reaction type is unknown.","url_abs":"https://arxiv.org/abs/2007.13437v2","url_pdf":"https://arxiv.org/pdf/2007.13437v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"drug-discovery","task_name":"Drug Discovery"},{"task_slug":"retrosynthesis","task_name":"Retrosynthesis"},{"task_slug":"single-step-retrosynthesis","task_name":"Single-step retrosynthesis"}],"methods":[{"method_slug":"ebm","method_name":"EBM"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/single-step-retrosynthesis-on-uspto-50k","task":"Single-step retrosynthesis","dataset":"USPTO-50k","model":"Dual-TF (reaction class as prior)","rank_in_archive_order":7,"of":35,"metrics":{"Top-1 accuracy":"65.7","Top-10 accuracy":"85.9","Top-3 accuracy":"81.9","Top-5 accuracy":"84.7"},"uses_additional_data":false},{"leaderboard":"/sota/single-step-retrosynthesis-on-uspto-50k","task":"Single-step retrosynthesis","dataset":"USPTO-50k","model":"Dual-TF (reaction class unknown)","rank_in_archive_order":21,"of":35,"metrics":{"Top-1 accuracy":"53.6","Top-10 accuracy":"77.0","Top-3 accuracy":"70.7","Top-5 accuracy":"74.6"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2007.13437","atlas_url":"https://app.syntology.ai/?focus=2007.13437","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}