# Closing the Prior-Posterior Loop: Self-Reflective Molecular Design with Analysis-Driven LLM Iteration

arXiv: https://arxiv.org/abs/2606.09520 · HTML: https://mlsanity.com/paper/2606.09520/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 39 references that are in the map.
- First posted on arXiv 8 Jun 2026.

## Builds on (most cited first, 6 of 6)

- [Directional Message Passing for Molecular Graphs](https://mlsanity.com/paper/2003.03123/) (2020)
- [Inverse design of 3d molecular structures with conditional generative neural networks](https://mlsanity.com/paper/2109.04824/) (2021)
- [Informing Geometric Deep Learning with Electronic Interactions to Accelerate Quantum Chemistry](https://mlsanity.com/paper/2105.14655/) (2021)
- [Graph Neural Networks for Molecules](https://mlsanity.com/paper/2209.05582/) (2022)
- [Structured Scaling of AI Discovery Across Diverse Scientific Domains](https://mlsanity.com/paper/2604.19341/) (2026)
- [DenseSteer: Steering Small Language Models towards Dense Math Reasoning](https://mlsanity.com/paper/2605.29247/) (2026)

## Abstract (arXiv)

Can a general-purpose large language model design molecules with the precision of a seasoned chemist? Current LLM-based frameworks answer this question with scalar feedback loops - generate, score, reject - that amount to informed trial-and-error. Here we show that replacing a single number with the full physicochemical rationale from first-principles calculations transforms the LLM from a stochastic sampler into a causal reasoner. Our system couples retrieval-augmented generation with a self-reflection module that feeds orbital energies, atomic charges, and electron densities - rather than compressed scores - back into the design loop. On HOMO-LUMO gap targets from 2.0 to 5.0 eV, this structure-property-relationship (SPR) reflection achieves a deviation as low as 0.0014 eV with a 100% success rate under the SPR+RAG configuration, consistently outperforming scalar-feedback and non-reflective baselines in median and mean deviation. The framework generalizes seamlessly to dipole-moment design, synthetic accessibility optimization, and molecular docking, and proves robust across 7 distinct LLM backbones. These results establish a new paradigm: when the model understands not only that a molecule fails, but why, iterative molecular design becomes genuinely mechanistic.