Reinforced Generation of Combinatorial Structures: Ramsey Numbers
Not on the map. Why?
Abstract (arXiv)
We present improved lower bounds for nine classical Ramsey numbers: R(3, 13) is increased from 60 to 61, R(3, 18) from 99 to 100, R(4, 13) from 138 to 139, R(4, 14) from 147 to 148, R(4, 15) from 158 to 159, R(4, 16) from 170 to 174, R(4, 18) from 205 to 209, R(4, 19) from 213 to 219, and R(4, 20) from 234 to 237. These results were achieved using AlphaEvolve, an LLM-based code mutation agent. Beyond these new results, we successfully recovered lower bounds for all Ramsey numbers known to be exact, and matched the best known lower bounds across many other cases. These include bounds for which previous work does not detail the algorithms used. Virtually all known Ramsey lower bounds are derived computationally, with bespoke search algorithms each delivering a handful of results. AlphaEvolve is a single meta-algorithm yielding search algorithms for all of our results.
Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 6 of its 15 references that are in the map. First posted on arXiv 10 Mar 2026.
Builds on
6 of 15 references.
References · 6
- 2025AlphaEvolve: A coding agent for scientific and algorithmic discoverycited by 763
- 2025Mathematical exploration and discovery at scalecited by 66
- 2021Constructions in combinatorics via neural networkscited by 46
- 2024PatternBoost: Constructions in Mathematics with a Little Help from AIcited by 32
- 2025Reinforced Generation of Combinatorial Structures: Hardness of Approximationcited by 8
- 2024A Formal Proof of R(4,5)=25cited by 1
Built on by
Citing papers per year: 2026: 8.
8 papers.
- 2026New Bounds for Zarankiewicz Numbers via Reinforced LLM Evolutionary Searchcited by 4
- 2026pAI/MSc: ML Theory Research with Humans on the Loopcited by 2
- 2026ScienceIDE: Turning World's Scientific Codebase into Agent Learnable Environmentscited by 2
- 2026k-server-bench: Automating Potential Discovery for the k-Server Conjecturecited by 0
- 2026Genetic Algorithms with Optimization Guided Operatorscited by 0
- 2026Stable Menus of Public Goods: AI-Enabled Progresscited by 0
- 2026EvoMem: Memory-Augmented Evolution for Code Optimizationcited by 0
- 2026RamseyGadgets: A Graph Construction Dataset for LLMscited by 0
- 2026ScienceIDE: Turning World's Scientific Codebase into Agent Learnable Environments16 Sep 2026
- 2026RamseyGadgets: A Graph Construction Dataset for LLMs15 Aug 2026
- 2026EvoMem: Memory-Augmented Evolution for Code Optimization11 Aug 2026
- 2026Stable Menus of Public Goods: AI-Enabled Progress15 Jun 2026
- 2026Genetic Algorithms with Optimization Guided Operators10 Jun 2026
- 2026New Bounds for Zarankiewicz Numbers via Reinforced LLM Evolutionary Search1 May 2026
- 2026pAI/MSc: ML Theory Research with Humans on the Loop22 Apr 2026
- 2026k-server-bench: Automating Potential Discovery for the k-Server Conjecture8 Apr 2026