EVENT DOSSIER
The researchers used an iterative evolution optimization algorithm for large language models instead of direct solving, which significantly improved the optimal solution for the known best radius of the circle in the Packomania csqv benchmark. This method increased the performance of the 10 known best solutions ranging from 101 to 114 by 2.4% to 5.4%, with the entire process taking 15 iterations. The total cost of the LLM was 27.72 dollars. During the validation phase, an independent verifier scored the candidate solutions, retaining those that improved and eliminating those that failed. Packomania has independently accepted these results. The related paper is published on arxiv.org/abs/2609.05093, and the code and solutions can be found at github.com/ucsandman/discovery-loop. The benchmark data is available at packomania.com/csqv/csqv.html.
作者利用大语言模型（LLM）迭代演化优化算法，而非直接求解打包问题。在 Packomania csqv 基准测试中，该方法将 N 从 101 到 114 的 10 个已知最佳圆半径和的最优解提升了 2.4% 至 5.4%，共耗时 15 次迭代。整个过程由独立验证器对候选方案进行评分，保留改进并剔除失败案例，总 LLM 成本为 27.72 美元。Packomania 已独立接受这些结果。相关论文发布于 arxiv.org/abs/2609.05093，代码及解决方案位于 github.com/ucsandman/discovery-loop，基准测试数据在 packomania.com/csqv/csqv.html。