Research map / Learning theory and algorithms / Matrix and tensor methods
RECIPE: Learning to Rank Complete Precursor Sets for Inorganic Retrosynthesis
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes RECIPE, a method that learns to rank complete precursor sets (rather than individual precursors) for single-step inorganic retrosynthesis, addressing unknown set size.
Problem
Single-step inorganic retrosynthesis is scored by recovering the complete precursor set, but data-driven systems rank individual precursors and use threshold or count heuristics to assemble sets, producing incomplete, over-complete, or near-miss sets when the true set size is unknown.
Approach
Reformulates precursor prediction as ranking complete precursor sets directly, so set-level plausibility is learned instead of assembled from per-precursor scores with heuristics.
Closest papers by meaning
- Retrieval-Retro: Retrieval-based Inorganic Retrosynthesis with Expert Knowledge (NeurIPS 2024 · similarity 0.92)
- RetroBridge: Modeling Retrosynthesis with Markov Bridges (ICLR 2024 · similarity 0.87)
- Advancing Retrosynthesis with Retrieval-Augmented Graph Generation (AAAI 2025 · similarity 0.87)
- LLM-Augmented Chemical Synthesis and Design Decision Programs (ICML 2025 · similarity 0.87)
- URSA: Chemistry-Aware Benchmark for Utilitarian Retrosynthesis Assessment (NeurIPS 2026 · similarity 0.86)
- RetroOOD: Understanding Out-of-Distribution Generalization in Retrosynthesis Prediction (AAAI 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.