maze-llm-benchmark-v1
B待确认Benchmark推理游戏推理开放mit
发布方:erdem-erdem
热度5.7▲ 0.1
下载量 · 30天
17
Hugging Face
GitHub Stars
—
代码仓库
论文被引
—
Semantic Scholar
跑分模型 · 30天
—
Leaderboard results
简介
Maze LLM Benchmark (multi-size × multi-algorithm) This dataset was generated with the maze-dataset library using the LatticeMazeGenerators algorithms (DFS, Wilson, Percolation, DFS+Percolation). Each example includes an ASCII maze (map) without solution marks, the generation algorithm, grid_n, the shortest-path path_length, and the NESW moves string of the shortest path. Grid sizes: [5, 7, 9, 11, 13]Algorithms: ['dfs', 'wilson', 'recdiv', 'percolation', 'dfs_percolation']Examples… See the full description on the dataset page: https://huggingface.co/datasets/erdem-erdem/maze-llm-benchmark-v1.