heurigen/heurigen-data
收藏Hugging Face2025-05-22 更新2025-10-25 收录
下载链接:
https://hf-mirror.com/datasets/heurigen/heurigen-data
下载链接
链接失效反馈官方服务:
资源简介:
HeuriGen是一个用于严格评估大型语言模型在组合优化问题上的基准和代理评估框架。该框架通过引入具有明确客观目标和广泛解决方案空间的现实世界组合优化任务,要求模型具备创造性算法设计、多步骤规划、工具使用和适应性推理的能力。
HeuriGen is a benchmark and agentic evaluation framework designed to rigorously assess Large Language Models (LLMs) on combinatorial optimization (CO) problems, requiring creative algorithm design, multi-step planning, tool use, and adaptive reasoning capabilities through real-world CO tasks with well-defined objectives and expansive solution spaces.
提供机构:
heurigen



