Luo Mai

The paper proposes FeasiGen, a method to automatically create infeasible tasks for tool-using agents, and finds that most current agents struggle significantly to detect and stop when faced with such impossible tasks.

The paper introduces Ryze, an automated system that synthesizes evidence-enriched Question-Answering (QA) pairs from raw biomedical papers, resulting in a specialized VLM (BioVLM-8B) that significantly outperforms existing models on biomedical benchmarks.