15 papers
When Reasoning Is Not Enough: Proactive Inquiry for Cross-Modal Safety in Multimodal Large Language Models
Wei Cai, Jian Zhao, Yuchen Yuan +1
Multimodal large language models (MLLMs) can act unsafely when individually benign images and texts are interpreted jointly. Beyond implicit reasoning risk, where reasoning drifts…
CaLR: Causal Latent Revision for Robust Diffusion Reasoning
Wei Cai, Jian Zhao, Yuchen Yuan +1
Autoregressive (AR) models suffer from local greediness, while diffusion language models (DLMs) often lack the strict causal structure required for reasoning. To combine the advant…
AeroCopilotBench: Safety-Gated Evaluation of LLM Agents on Aircraft Emergency Procedures in an Executable Cockpit
Yuchen Yuan, Zhenghuang Wu, Yuangan Li +2
Aviation knowledge question answering cannot directly assess the operational effectiveness and safety compliance of large language models throughout aircraft emergency procedures.…
Abstention as an Action Can Kill Both the Reward Gradient and the KL Anchor: Collapse Law and Repair for Error-Penalized Reinforcement Learning
Xujun Che, Yuchen Yuan, Weida Zhao +1
Error-penalized scoring rules ( for a correct answer, for a wrong one, for abstaining) are increasingly prescribed against hallucination: a rational agent facing such…
Metis: Learning to Jailbreak LLMs via Self-Evolving Metacognitive Policy Optimization
Huilin Zhou, Jian Zhao, Yilu Zhong +7
Red teaming is critical for uncovering vulnerabilities in Large Language Models (LLMs). While automated methods have improved scalability, existing approaches often rely on static…
The RoboSense Challenge: Sense Anything, Navigate Anywhere, Adapt Across Platforms
Lingdong Kong, Shaoyuan Xie, Zeying Gong +135
Autonomous systems are increasingly deployed in open and dynamic environments -- from city streets to aerial and indoor spaces -- where perception models must remain reliable under…
Aetheria: A multimodal interpretable content safety framework based on multi-agent debate and collaboration
Yuxiang He, Jian Zhao, Yuchen Yuan +8
The exponential growth of digital content presents significant challenges for content safety. Current moderation systems, often based on single models or fixed pipelines, exhibit l…
TeleAI-Safety: A comprehensive LLM jailbreaking benchmark towards attacks, defenses, and evaluations
Xiuyuan Chen, Jian Zhao, Yuxiang He +10
While the deployment of large language models (LLMs) in high-value industries continues to expand, the systematic assessment of their safety against jailbreak and prompt-based atta…
A Parameter-Efficient Mixture-of-Experts Framework for Cross-Modal Geo-Localization
LinFeng Li, Jian Zhao, Zepeng Yang +6
We present a winning solution to RoboSense 2025 Track 4: Cross-Modal Drone Navigation. The task retrieves the most relevant geo-referenced image from a large multi-platform corpus…
RADAR: A Risk-Aware Dynamic Multi-Agent Framework for LLM Safety Evaluation via Role-Specialized Collaboration
Xiuyuan Chen, Jian Zhao, Yuchen Yuan +8
Existing safety evaluation methods for large language models (LLMs) suffer from inherent limitations, including evaluator bias and detection failures arising from model homogeneity…
Visual Attention Reasoning via Hierarchical Search and Self-Verification
Wei Cai, Jian Zhao, Yuchen Yuan +4
Multimodal Large Language Models (MLLMs) frequently hallucinate due to their reliance on fragile, linear reasoning and weak visual grounding. We propose Visual Attention Reasoning…
OrthAlign: Orthogonal Subspace Decomposition for Non-Interfering Multi-Objective Alignment
Liang Lin, Zhihao Xu, Junhao Dong +10
Large language model (LLM) alignment faces a critical dilemma when addressing multiple human preferences: improvements in one dimension frequently come at the expense of others, cr…
When Safe Unimodal Inputs Collide: Optimizing Reasoning Chains for Cross-Modal Safety in Multimodal Large Language Models
Wei Cai, Shujuan Liu, Jian Zhao +6
Multimodal Large Language Models (MLLMs) are susceptible to the implicit reasoning risk, wherein innocuous unimodal inputs synergistically assemble into risky multimodal data that…
ERF-BA-TFD+: A Multimodal Model for Audio-Visual Deepfake Detection
Xin Zhang, Jiaming Chu, Jian Zhao +5
Deepfake detection is a critical task in identifying manipulated multimedia content. In real-world scenarios, deepfake content can manifest across multiple modalities, including au…
Never compromise with vulnerabilities: a comprehensive survey on AI governance
Yuchu Jiang, Jian Zhao, Yuchen Yuan +62
The rapid advancement of AI has expanded its capabilities across domains, yet introduced critical technical vulnerabilities, such as algorithmic bias and adversarial sensitivity, t…