Zixuan Ke

LiveResearchBench: A Live Benchmark for User-Centric Deep Research in the Wild

WAFER-QA: Evaluating Vulnerabilities of Agentic Workflows with Agent-as-Judge

MAS-Zero: Designing Multi-Agent Systems with Zero Supervision

Beyond Accuracy: Dissecting Mathematical Reasoning for LLMs Under Reinforcement Learning

FaithEval: Can Your Language Model Stay Faithful to Context, Even If "The Moon is Made of Marshmallows"

Expanding the Web, Smaller Is Better: A Comprehensive Study in Post-training