DeltaBench

Overview 🌐 Website • 🤗 Hugging Face • ⏬ Data • 📃 Paper • 🖥️ Github 💥 DeltaBench DeltaBench the first dataset to analyze the qualities of the long CoTs generated by o1-like models and evaluate the critique abilities to Detect Error in Long CoT ReAsoning of existing critic models and PRMs. Specifically, DeltaBench c

Papers using DeltaBench (2)

Can Large Language Models Detect Errors in Long Chain-of-Thought Reasoning?2025SPC: Evolving Self-Play Critic via Adversarial Games for LLM Reasoning2025