{"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/a-self-feedback-knowledge-elicitation","title":"A Self-feedback Knowledge Elicitation Approach for Chemical Reaction Predictions","arxiv_id":"2404.09606","date":"2024-04-15","proceeding":null,"authors":["PengFei Liu","Jun Tao","Zhixiang Ren"],"abstract":"The task of chemical reaction predictions (CRPs) plays a pivotal role in advancing drug discovery and material science. However, its effectiveness is constrained by the vast and uncertain chemical reaction space and challenges in capturing reaction selectivity, particularly due to existing methods' limitations in exploiting the data's inherent knowledge. To address these challenges, we introduce a data-curated self-feedback knowledge elicitation approach. This method starts from iterative optimization of molecular representations and facilitates the extraction of knowledge on chemical reaction types (RTs). Then, we employ adaptive prompt learning to infuse the prior knowledge into the large language model (LLM). As a result, we achieve significant enhancements: a 14.2% increase in retrosynthesis prediction accuracy, a 74.2% rise in reagent prediction accuracy, and an expansion in the model's capability for handling multi-task chemical reactions. This research offers a novel paradigm for knowledge elicitation in scientific research and showcases the untapped potential of LLMs in CRPs.","url_abs":"https://arxiv.org/abs/2404.09606v1","url_pdf":"https://arxiv.org/pdf/2404.09606v1.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":[{"paper_slug":"a-self-feedback-knowledge-elicitation","repo_url":"https://github.com/ai-hpc-research-team/slm4crp","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"chemical-reaction-prediction","task_name":"Chemical Reaction Prediction"},{"task_slug":"drug-discovery","task_name":"Drug Discovery"},{"task_slug":"forward-reaction-prediction","task_name":"Forward reaction prediction"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"large-language-model","task_name":"Large Language Model"},{"task_slug":"prompt-learning","task_name":"Prompt Learning"},{"task_slug":"reagent-prediction","task_name":"Reagent Prediction"},{"task_slug":"retrosynthesis","task_name":"Retrosynthesis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/chemical-reaction-prediction-on-mol","task":"Chemical Reaction Prediction","dataset":"Mol-Instruction","model":"SLM4CRP","rank_in_archive_order":1,"of":1,"metrics":{"Exact":"0.674","METEOR":"0.901","Morgan FTS":"0.854","Validity":"0.998"},"uses_additional_data":false},{"leaderboard":"/sota/forward-reaction-prediction-on-mol","task":"Forward reaction prediction","dataset":"Mol-Instruction","model":"SLM4CRP","rank_in_archive_order":1,"of":2,"metrics":{"Exact":"0.945","METEOR":"0.993","Morgan FTS":"0.986","Validity":"0.997"},"uses_additional_data":false},{"leaderboard":"/sota/reagent-prediction-on-mol-instruction","task":"Reagent Prediction","dataset":"Mol-Instruction","model":"SLM4CRP","rank_in_archive_order":1,"of":2,"metrics":{"Exact":"0.284","METEOR":"0.744","Morgan FTS":"0.649","Validity":"1"},"uses_additional_data":false},{"leaderboard":"/sota/retrosynthesis-on-mol-instruction","task":"Retrosynthesis","dataset":"Mol-Instruction","model":"SLM4CRP","rank_in_archive_order":1,"of":2,"metrics":{"Exact":"0.757","METEOR":"0.95","Morgan FTS":"0.905","Validity":"0.994"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}