7 papers
ToolHazard: Scaling Adversarial Environments for Security Evaluation and Alignment of LLM-based Agents
Yutao Mou, Pengfei Yang, Zhe Yin +6
Large language model (LLM) agents integrated with external tools are vulnerable to indirect prompt injections embedded in environmental states. However, existing studies largely re…
SecureWebArena: A Holistic Security Evaluation Benchmark for LVLM-based Web Agents
Zonghao Ying, Yangguang Shao, Jianle Gan +8
Large vision-language model (LVLM)-based web agents are emerging as powerful tools for automating complex online tasks. However, when deployed in real-world environments, they face…
Select-then-Solve: Paradigm Routing as Inference-Time Optimization for LLM Agents
Heng Zhou, Zelin Tan, Zhemeng Zhang +15
When an LLM-based agent improves on a task, is the gain from the model itself or from the reasoning paradigm wrapped around it? We study this question by comparing six inference-ti…
TodoEvolve: Learning to Architect Agent Planning Systems
Jiaxi Liu, Yanzuo Jiang, Guibin Zhang +5
Planning has become a central capability for contemporary agent systems in navigating complex, long-horizon tasks, yet existing approaches predominantly rely on fixed, hand-crafted…
TermiGen: High-Fidelity Environment and Robust Trajectory Synthesis for Terminal Agents
Kaijie Zhu, Yuzhou Nie, Yijiang Li +10
Executing complex terminal tasks remains a significant challenge for open-weight LLMs, constrained by two fundamental limitations. First, high-fidelity, executable training environ…
RoboSafe: Safeguarding Embodied Agents via Executable Safety Logic
Le Wang, Zonghao Ying, Xiao Yang +7
Embodied agents powered by vision-language models (VLMs) are increasingly capable of executing complex real-world tasks, yet they remain vulnerable to hazardous instructions that m…