ReTRACE: Feedback-Guided LLM Agent in a Chemistry Harness for Retrosynthesis Planning

Key points

• Computer-aided retrosynthesis has advanced through single-step prediction, tree search and route scoring, but complex targets still require judgement over the whole route, where stereochemistry, functional-group compatibility, cost and greenness interact.1
• Routes are judged by a new metric that combines programmatic checks with adversarial LLM review, and its verdicts are compared with a synthetic chemist's.1
• The agent learns continually from evaluator feedback during search, diagnosing weak routes and backtracking to revise them.1
• Two prospective wet-lab syntheses show that ReTRACE route proposals can be converted into executable procedures after expert revision.1

Sources (1)

• [1]ReTRACE: Feedback-Guided LLM Agent in a Chemistry Harness for Retrosynthesis Planning
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 17, 03:22 PM

  Here we present \textbf{\mbox{ReTRACE}} (REtrosynthesis with Tool-orchestrated Reasoning And Context-aware Evidence), an agent that places a large language model (LLM) in a chemistry harness that integrates molecular tools, complementary single-step prediction engines, a purchasable starting-material library, a route memory and a multi-criterion route evaluator.

  Computer-aided retrosynthesis has advanced through single-step prediction, tree search and route scoring, but complex targets still require judgement over the whole route, where stereochemistry, functional-group compatibility, cost and greenness interact.

Extractive summary: sentences quoted from the sources.