# Atom-anchored LLMs speak Chemistry: A Retrosynthesis Demonstration

arXiv: https://arxiv.org/abs/2510.16590 · HTML: https://mlsanity.com/paper/2510.16590/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 54 references that are in the map.
- First posted on arXiv 18 Oct 2025.

## Builds on (most cited first, 20 of 20)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [ChemCrow: Augmenting large-language models with chemistry tools](https://mlsanity.com/paper/2304.05376/) (2023)
- [Large-Scale Chemical Language Representations Capture Molecular Structure and Properties](https://mlsanity.com/paper/2106.09553/) (2021)
- [What can Large Language Models do in chemistry? A comprehensive benchmark on eight tasks](https://mlsanity.com/paper/2305.18365/) (2023)
- [Learning to Plan Chemical Syntheses](https://mlsanity.com/paper/1708.04202/) (2017)
- [Learning Graph Models for Retrosynthesis Prediction](https://mlsanity.com/paper/2006.07038/) (2020)
- [Training a Scientific Reasoning Model for Chemistry](https://mlsanity.com/paper/2506.17238/) (2025)
- [Beyond Chemical QA: Evaluating LLM's Chemical Reasoning with Modular Chemical Operations](https://mlsanity.com/paper/2505.21318/) (2025)
- [Can Large Language Models Empower Molecular Property Prediction?](https://mlsanity.com/paper/2307.07443/) (2023)
- [RetroBridge: Modeling Retrosynthesis with Markov Bridges](https://mlsanity.com/paper/2308.16212/) (2023)
- [Structured Chemistry Reasoning with Large Language Models](https://mlsanity.com/paper/2311.09656/) (2023)
- [Chemical reasoning in LLMs unlocks strategy-aware synthesis planning and reaction mechanism elucidation](https://mlsanity.com/paper/2503.08537/) (2025)
- [Models Matter: The Impact of Single-Step Retrosynthesis on Synthesis Planning](https://mlsanity.com/paper/2308.05522/) (2023)
- [SmileyLlama: Modifying Large Language Models for Directed Chemical Space Exploration](https://mlsanity.com/paper/2409.02231/) (2024)
- [Can Large Language Models Understand Molecules?](https://mlsanity.com/paper/2402.00024/) (2024)
- [BatGPT-Chem: A Foundation Large Model For Retrosynthesis Prediction](https://mlsanity.com/paper/2408.10285/) (2024)

## Built on by (most cited first, 4 of 4)

- [Synthelite: Chemist-aligned and feasibility-aware synthesis planning with LLMs](https://mlsanity.com/paper/2512.16424/) (2025)
- [When Single Answer Is Not Enough: Rethinking Single-Step Retrosynthesis Benchmarks for LLMs](https://mlsanity.com/paper/2602.03554/) (2026)
- [Agentic generation of verifiable rules for deterministic, self-expanding reaction classification](https://mlsanity.com/paper/2607.01061/) (2026)
- [Reinforced Reasoning for End-to-End Retrosynthetic Planning](https://mlsanity.com/paper/2603.29723/) (2026)

## Abstract (arXiv)

Applications of machine learning in chemistry are often limited by the scarcity and expense of labeled data, restricting traditional supervised methods. In this work, we introduce a framework for molecular reasoning using general-purpose Large Language Models (LLMs) that operates without requiring task-specific model training. Our method anchors chain-of-thought reasoning to the molecular structure by using unique atomic identifiers. First, the LLM performs a zero-shot task to identify relevant fragments and their associated chemical labels or transformation classes. In an optional second step, this position-aware information is used in a few-shot task with provided class examples to predict the chemical transformation. We apply our framework to single-step retrosynthesis, a task where LLMs have previously underperformed. Across academic benchmarks and expert-validated drug discovery molecules, our work enables LLMs to achieve high success rates in identifying chemically plausible reaction sites (≥90%), named reaction classes (≥40%), and final reactants (≥74%). Ultimately, our work establishes a general blueprint for applying LLMs to challenges where molecular reasoning and molecular transformations are key, positioning atom-anchored LLMs as a powerful solution for data-scarce chemistry domains.