Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents Paper • 2602.13379 • Published Feb 13 • 4
Tool Specifications Matter: Uncovering and Mitigating Safety Risks in AI Agents Paper • 2607.29254 • Published Jul 31 • 2
DataFlex: A Unified Framework for Data-Centric Dynamic Training of Large Language Models Paper • 2603.26164 • Published Mar 27 • 280