12 papers
Self-Evolving Just-In-Time Memory for Proactive Embodied Safety
Bingrui Sima, Lizhong Wang, Xiaoya Lu +2
While Vision-Language Models (VLMs) have empowered embodied agents to execute complex household tasks, they struggle to proactively handle dynamically emerging hazards during close…
Exploring Agentic Tool-Calling Decisions via Uncertainty-Aligned Reinforcement Learning
Yijin Zhou, Linqian Zeng, Xiaoya Lu +4
Large language model (LLM)-based agents often make suboptimal tool-use decisions, including unsupported tool invocation and hallucinated direct responses, which may accumulate erro…
Valence-Arousal Subspace in LLMs: Circular Emotion Geometry and Multi-Behavioral Control
Lihao Sun, Lewen Yan, Xiaoya Lu +3
We show that emotion vectors in LLMs are organized by a two-dimensional valence-arousal (VA) subspace exhibiting circular geometry. Through principal component decomposition and ri…
LLMs know their vulnerabilities: Uncover Safety Gaps through Natural Distribution Shifts
Qibing Ren, Hao Li, Dongrui Liu +7
Safety concerns in large language models (LLMs) have gained significant attention due to their exposure to potentially harmful data during pre-training. In this paper, we identify…
HomeGuard: VLM-based Embodied Safeguard for Identifying Contextual Risk in Household Task
Xiaoya Lu, Yijin Zhou, Zeren Chen +6
Vision-Language Models (VLMs) empower embodied agents to execute complex instructions, yet they remain vulnerable to contextual safety risks where benign commands become hazardous…
INFA-Guard: Mitigating Malicious Propagation via Infection-Aware Safeguarding in LLM-Based Multi-Agent Systems
Yijin Zhou, Xiaoya Lu, Dongrui Liu +2
The rapid advancement of Large Language Model (LLM)-based Multi-Agent Systems (MAS) has introduced significant security vulnerabilities, where malicious influence can propagate vir…