19 papers
HarmProfile: Characterizing Harmful Distributions in Frontier LLMs
Zhouyuan Ma, Yutao Wu, Hanxun Huang +6
Frontier large language models (LLMs) safety evaluation has largely treated harmful generation as an attack outcome rather than as an object of analysis. Consequently, little is kn…
UniLab: A Heterogeneous Architecture for Robot RL Beyond GPU-Dominant Paradigms
Yufei Jia, Zhanxiang Cao, Mingrui Yu +48
Simulation-based RL for contemporary robot control is increasingly organized around GPU-resident simulation: physics, rollout collection, and learning are placed on a single GPU-ce…
Constitutional On-Policy Safe Distillation
Ming Wen, Yuxuan Liu, Kun Yang +8
On-policy self-distillation (OPSD) has emerged as an efficient post-training paradigm by using a teacher conditioned on privileged information to provide dense token-level supervis…
AgentDoG 1.5: A Lightweight and Scalable Alignment Framework for AI Agent Safety and Security
Dongrui Liu, Yu Li, Zhonghao Yang +47
Modern open-world agents such as OpenClaw exhibit powerful cross-environment execution capabilities yet introduce broad new safety risk sources. Meanwhile, advanced frontier AI mod…
When Medical Safety Alignment Fails: A Benchmark for Evaluating LLMs on High-Risk Medical Queries
Yige Li, Jun Sun, Wei Zhao +5
Large language models (LLMs) are increasingly used for medical and health-related questions, yet their safety in high-risk medical scenarios remains poorly understood. We introduce…
Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses
Xiao Li, Xiang Zheng, Yifeng Gao +35
Embodied Artificial Intelligence (Embodied AI) integrates perception, cognition, planning, and interaction into agents that operate in open-world, safety-critical environments. As…
BabelSafe: A Policy-Grounded Multilingual Safety Benchmark for LLMs
Yunhan Zhao, Zhaorun Chen, Xingjun Ma +2
As Large Language Models (LLMs) are increasingly deployed in cross-linguistic contexts, ensuring safety across diverse regulatory and cultural environments has become a critical ch…
Steering the Verifiability of Multimodal AI Hallucinations
Jianhong Pang, Ruoxi Cheng, Ziyi Ye +4
AI applications driven by multimodal large language models (MLLMs) are prone to hallucinations and pose considerable risks to human users. Crucially, such hallucinations are not eq…
Agent4POI: Agentic Context-Conditioned Affordance Reasoning for Multimodal Point-of-Interest Recommendation
Jinze Wang, Yangchen Zeng, Tiehua Zhang +5
We introduce Agent4POI, the first POI recommendation framework that generates context-conditioned multimodal representations at recommendation time, rather than relying on static P…
AgentHazard: A Benchmark for Evaluating Harmful Behavior in Computer-Use Agents
Yunhao Feng, Yifan Ding, Yingshui Tan +6
Computer-use agents extend language models from text generation to persistent action over tools, files, and execution environments. Unlike chat systems, they maintain state across…
What Does a Sharing Question Add? Auditing LLM Survey Scores for Misinformation
Zonghuan Xu, Xiang Zheng, Yutao Wu +1
Evaluating misinformation requires distinguishing whether readers believe content from whether they would share it. Asking large language models (LLMs) both questions yields two sc…
Attention in Space: Functional Roles of VLM Heads for Spatial Reasoning
Xueqi Ma, Shuo Yang, Yanbei Jiang +6
Despite remarkable advances in large Vision-Language Models (VLMs), spatial reasoning remains a persistent challenge. In this work, we investigate how attention heads within VLMs c…
OOD-MMSafe: Advancing MLLM Safety from Harmful Intent to Hidden Consequences
Ming Wen, Kun Yang, Jingyu Zhang +4
While safety alignment for Multimodal Large Language Models (MLLMs) has gained significant attention, current paradigms primarily target malicious intent or situational violations.…
Backdoor4Good: Benchmarking Beneficial Uses of Backdoors in LLMs
Yige Li, Wei Zhao, Zhe Li +6
Backdoor mechanisms have traditionally been studied as security threats that compromise the integrity of machine learning models. However, the same mechanism -- the conditional act…
Internal Safety Collapse in Frontier Large Language Models
Yutao Wu, Xiao Liu, Yifeng Gao +7
This work identifies a critical failure mode in frontier large language models (LLMs), which we term Internal Safety Collapse (ISC): under certain task conditions, models enter a s…
BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based Agents
Yunhao Feng, Yige Li, Yutao Wu +6
Large language model (LLM) agents execute tasks through multi-step workflows that combine planning, memory, and tool use. While this design enables autonomy, it also expands the at…
Just Ask: Curious Code Agents Reveal System Prompts in Frontier LLMs
Xiang Zheng, Yutao Wu, Hanxun Huang +5
Autonomous code agents built on large language models are reshaping software and AI development through tool use, long-horizon reasoning, and self-directed interaction. However, th…
PaperAsk: A Benchmark for Reliability Evaluation of LLMs in Paper Search and Reading
Yutao Wu, Xiao Liu, Yunhao Feng +2
Large Language Models (LLMs) increasingly serve as research assistants, yet their reliability in scholarly tasks remains under-evaluated. In this work, we introduce PaperAsk, a ben…
RedDiffuser: Auditing Multimodal Safety Failures in Vision-Language Models via Reinforced Diffusion
Ruofan Wang, Xingjun Ma
Large Vision-Language Models (VLMs) are increasingly deployed in open-ended environments, where ensuring reliable safety under multimodal inputs is critical. However, existing eval…