Search-QA

We publicly release a new large-scale dataset, called SearchQA, for machine comprehension, or question-answering. Unlike recently released datasets, such as DeepMind CNN/DailyMail and SQuAD, the proposed SearchQA was constructed to reflect a full pipeline of general question-answering. That is, we start not from an exi

Papers using Search-QA (5)

StepOPSD: Step-Aware Online Preference Distillation for Agent Reinforcement Learning2026T$^2$PO: Uncertainty-Guided Exploration Control for Stable Multi-Turn Agentic Reinforcement Learning2026Dynamic Skill Lifecycle Management for Agentic Reinforcement Learning2026Self-Distilled Agentic Reinforcement Learning2026SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization2026