HMMT 25

The 'HMMT 25' dataset/benchmark contains a collection of challenging mathematical problems used to evaluate the reasoning capabilities of large language models (LLMs).

Papers using HMMT 25 (3)

PRISM: Pushing the Frontier of Deep Think via Process Reward Model-Guided Inference2026PromptCoT 2.0: Scaling Prompt Synthesis for Large Language Model Reasoning2025PromptCoT 2.0: Scaling Prompt Synthesis for Large Language Model Reasoning2025