Research map / Learning theory and algorithms / Combinatorial and search optimization
Retro*: Learning Retrosynthetic Planning with Neural Guided A* Search
Binghong Chen, Chengtao Li, Hanjun Dai, Le Song
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Learn molecule synthesis costs from prior planning experience and use them to guide AND-OR best-first search (ablation) | Retro* · USPTO test route dataset | Success rate: 79.47% -> 86.84% | Improved |
| Replace the learned molecule value function with zero (ablation) | Retro* · USPTO test route dataset | One-step model calls: 156.58 -> 208.58 | Hurt |
| Use the learned value function instead of rollout reward estimation | MCTS · USPTO test route dataset | Success rate: 33.68% -> 35.79% | Improved |
| Initialize DFPN-E proof numbers with the learned value function | DFPN-E · USPTO test route dataset | Success rate: 55.26% -> 53.68% | Hurt |
| Use the learned value function to initialize DFPN-E proof numbers | DFPN-E · USPTO test route dataset | One-step model calls: 279.67 -> 289.42 | Hurt |
Later papers that compared with this method
3 later accepted papers report Retro* as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- From Feasible to Practical: Pareto-Optimal Synthesis Planning (ICML 2026 · better on 11 of 13 shared results)
- Double-Ended Synthesis Planning with Goal-Constrained Bidirectional Search (NeurIPS 2024 · better on 8 of 9 shared results)
- SeeA*: Efficient Exploration-Enhanced A* Search by Selective Sampling (NeurIPS 2024 · better on 5 of 5 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Self-Improved Retrosynthetic Planning (ICML 2021 · similarity 0.95)
- Retrieval-Retro: Retrieval-based Inorganic Retrosynthesis with Expert Knowledge (NeurIPS 2024 · similarity 0.91)
- LLM-Augmented Chemical Synthesis and Design Decision Programs (ICML 2025 · similarity 0.90)
- Retro-fallback: retrosynthetic planning in an uncertain world (ICLR 2024 · similarity 0.90)
- RetroBridge: Modeling Retrosynthesis with Markov Bridges (ICLR 2024 · similarity 0.89)
- RETRO SYNFLOW: Discrete Flow-Matching for Accurate and Diverse Single-Step Retrosynthesis (NeurIPS 2025 · similarity 0.89)
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.
