How2Everything: Mining the Web for How-To Procedures to Evaluate and Improve LLMs
Open paper
Benchmark
Model
Rank
Results
mathematical-reasoning-on-aime24
Qwen3-8B (Instruct) + RL
#67
Acc: 59.06
multi-task-language-understanding-on-mmlu-pro
Qwen3-8B (Instruct) + RL
#87
Accuracy: 63.11
Rank counts only results with a code link.