Xuanliang Zhang

FinSearchComp: Towards a Realistic, Expert-Level Evaluation of Financial Search and Reasoning

Bounds of Chain-of-Thought Robustness: Reasoning Steps, Embed Norms, and Beyond

DiscoX: Benchmarking Discourse-Level Translation in Expert Domains

Learning-to-Context Slope: Evaluating In-Context Learning Effectiveness Beyond Performance Illusions

Beyond Score: A Multi-Agent System to Discover Capability and Behavioral Weaknesses in LLMs

MME-CC: A Challenging Multi-Modal Evaluation Benchmark of Cognitive Capacity

In-Context Transfer Learning: Demonstration Synthesis by Transferring Similar Tasks