Agentic Neurosymbolic Collaboration for Mathematical Discovery: A Case Study in Combinatorial Design

TL;DR

This paper demonstrates how a neurosymbolic AI system, combining large language models, symbolic tools, and human guidance, can collaboratively make a genuine mathematical discovery in combinatorial design, specifically establishing a new lower bound for Latin squares.

Contribution

It introduces a novel neurosymbolic framework for mathematical discovery, integrating AI, symbolic computation, and human insight, leading to a new theoretical bound in combinatorics.

Findings

AI effectively uncovered hidden structures and generated hypotheses.

Human steering was crucial for transforming dead ends into productive research.

The discovered lower bound was formally verified in Lean 4.

Abstract

We study mathematical discovery through the lens of neurosymbolic reasoning, where an AI agent powered by a large language model (LLM), coupled with symbolic computation tools, and human strategic direction, jointly produced a new result in combinatorial design theory. The main result of this human-AI collaboration is a tight lower bound on the imbalance of Latin squares for the notoriously difficult case $n \equiv 1 \pmod{3}$. We reconstruct the discovery process from detailed interaction logs spanning multiple sessions over several days and identify the distinct cognitive contributions of each component. The AI agent proved effective at uncovering hidden structure and generating hypotheses. The symbolic component consists of computer algebra, constraint solvers, and simulated annealing, which provides rigorous verification and exhaustive enumeration. Human steering supplied the critical research pivot that transformed a dead end into a productive inquiry. Our analysis reveals that multi-model deliberation among frontier LLMs proved reliable for criticism and error detection but unreliable for constructive claims. The resulting human-AI mathematical contribution, a tight lower bound of $4n(n{-}1)/9$, is achieved via a novel class of near-perfect permutations. The bound was formally verified in Lean 4. Our experiments show that neurosymbolic systems can indeed produce genuine discoveries in pure mathematics.