Large language model based agents increasingly rely on external tools to invoke services and interact with web resources. These tools can evolve through software updates, yet agentic frameworks may continue trusting them without re-verifying their safety after implementation changes. This tool update gap creates a supp...
Shi-Yu Kan, Yue Qi, Yu-Fei Han et al.· 0 citations
ADWM (Autoregressive Diffusion World Model), an evaluation framework that estimates the performance of a new LLM agent policy purely from pre-collected trajectories, achieves accurate value estimates and evaluation reliability across diverse multi-turn agent tasks, demonstrating its promise as a practical framework for...
Kai Liu, Guo-Jun Xiong, Wei-Nan Zhang et al.· arXiv.org· 1 citation
It is shown that a one-line guardrail achieves large single-shot ASR reductions, up to roughly 40 points, at near-zero over-refusal cost, which overstates deployed robustness by a systematic and predictable margin.
Haoxin An, Yunpeng Song, Zihao Bai et al.· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.