Transportation Safety and Impact Analysis

Latest papers

• Benchmarking Jailbreak Guardrails for Embodied Agents

  Xunguang Wang, Qingyue Wang, Yuguang Zhou, Zongjie Li, Wenxuan Wang, Shuai Wang · 6 October 2026

  Embodied agents powered by large language models and vision-language models are increasingly deployed in physical environments, but jailbreak attacks can induce these agents to perform physically harmful actions. A growing number of guardrail methods have been proposed to intercept dangerous behavio…

• Safety Reconstructed: Generative Modeling via Masked Diffusion Builds Strong Safety Guardrails

  Gert Lek, Abele Malan, Chaoyi Zhu, Pin-Yu Chen, Robert Birke, Lydia Chen · 29 September 2026

  Guard models are the last line of defense between a language model and a harmful output, yet their training objective is surprisingly narrow. Existing guards learn to predict a single verdict token from a conversational context, concentrating supervision on a single target. The consequences are stru…

• AdaGuard: An Adaptive Guard Model with User-defined Policies

  Yunhao Feng, Yifan Ding, Yuxiang Xie, Zheng Li, Mingrui Lao, Zeyuan Wang, Yanming Guo · 29 September 2026

  Guard models support the safe deployment of language model agents, but fixed risk taxonomies limit their ability to accommodate requirements that vary across applications and tasks. Under user-defined policies, detecting violations requires interpreting both the applicable rules and the agent's beha…

• InGuard: Towards Generalized Inner Guardrail for Safe Text-to-Image Generation

  Zeyu Wang, Xiaodan Li, Zhiwen Li, Yuefeng Chen, Hui Xue · 24 September 2026

  Modern text-to-image (T2I) models generate high-quality images from arbitrary user prompts, yet they can just as easily produce not-safe-for-work (NSFW) content. Conventional outer guardrails consist of two components: a prompt classifier that checks for risk before generation, and a post-hoc image …

• Visual Compliance via Executable Safety Rule Entailment

  Jisoo Kim (Sungkyunkwan University), TaeYoon Kwack (Sungkyunkwan University), Jinwoo Jang (Sungkyunkwan University), Honguk Woo (Sungkyunkwan University) · 17 September 2026

  Recent advances in LLMs and VLMs have enabled safety systems to reason beyond simple risk patterns toward more contextual and semantic safety concerns. However, as risk patterns continue to evolve and safety rules become more complex, existing training-based end-to-end safeguards face persistent cha…

• LongGuard: Mechanistic Analysis and Training-Free Mitigation of Long-Context Failure in Safety Guardrails

  Ziyang Chen, Xing Wu, Songlin Hu · 31 August 2026

  Safety guardrails serve as the last line of defense against harmful inputs and outputs of large language models (LLMs), yet they are trained and evaluated almost exclusively on short text. We present LongGuard, a framework that evaluates, mechanistically analyzes, and mitigates long-context guardrai…

• Reflex-Guard: A Low-Latency Guardrail for LLM Prompt Safety Using Dense Semantic Embeddings

  Istiaque Ahmed, Afia Anjum Borsha, Ranat Das Prangon, Abu-fuad Ahmad, Thi Hong Tran · 19 August 2026

  Large Language Models (LLMs) in real-world applications often face the risks of specially crafted prompts designed to bypass the safety controls. Existing guardrail methods, such as LLM-as-a-judge and cloud-based safety APIs are able to detect unsafe content. However, they often add a delay of about…

• ProbGuard: Calibrated Safety Risk Estimation from LLM Output Distributions

  Xinzhe Huang, Biwu Yao, Kedong Xiu, Mengnan Zhao, Di Wang, Puning Zhao, Tianhang Zheng · 12 August 2026

  Recent research on Large Language Model (LLM) safety has widely adopted guardrails to identify unsafe LLM outputs. Existing guardrails typically formulate safety assessment as a deterministic classification task, mapping a discrete token sequence to a discrete safety label. However, this paradigm ha…

• DreamGuard: Efficient Runtime Guardrail for LLM Agents via Risk-Aware World Model

  Wenhao Lin, Chenyu Yu, Xingwei Lin, Sicong Cao, Xiang Chen, Lei Xue, Le Yu, Letian Sha, Chunming Wu · 7 August 2026

  As large language model (LLM) agents increasingly invoke external tools and interact with real-world systems, unsafe actions may cause irreversible consequences on external states, user data, and downstream services. Recent runtime guardrails mitigate such risks by checking proposed actions before e…

• ReasoningGuard: Safeguarding Large Reasoning Models with Inference-time Safety Aha Moments

  Yuquan Wang, Mi Zhang, Yining Wang, Geng Hong, Mi Wen, Xiaoyu You, Min Yang · 7 May 2026

  Large Reasoning Models (LRMs) have demonstrated impressive performance in reasoning-intensive tasks, but they remain vulnerable to harmful content generation, particularly in the mid-to-late steps of their reasoning processes. Current defense methods, however, depend on costly fine-tuning and additi…

• Disentangled Safety Adapters Enable Efficient Guardrails and Flexible Inference-Time Alignment

  Kundan Krishna, Joseph Y Cheng, Charles Maalouf, Leon A Gatys · 4 May 2026

  Existing paradigms for ensuring AI safety, such as guardrail models and alignment training, often compromise either inference efficiency or development flexibility. We introduce Disentangled Safety Adapters (DSA), a novel framework addressing these challenges by decoupling safety-specific computatio…

• Safety Guardrails for LLM-Enabled Robots

  Zachary Ravichandran, Alexander Robey, Vijay Kumar, George J. Pappas, Hamed Hassani · 5 March 2026

  Although the integration of large language models (LLMs) into robotics has unlocked transformative capabilities, it has also introduced significant safety concerns, ranging from average-case LLM errors (e.g., hallucinations) to adversarial jailbreaking attacks, which can produce harmful robot behavi…

• Personalized Safety Alignment for Text-to-Image Diffusion Models

  Yu Lei, Jinbin Bai, Qingyu Shi, Aosong Feng, Hongcheng Gao, Xiao Zhang, Rex Ying · 6 February 2026

  Text-to-image diffusion models have revolutionized visual content generation, yet their deployment is hindered by a fundamental limitation: safety mechanisms enforce rigid, uniform standards that fail to reflect diverse user preferences shaped by age, culture, or personal beliefs. To address this, w…

• Let Them Down Easy! Contextual Effects of LLM Guardrails on User Perceptions and Preferences

  Mingqian Zheng, Wenjia Hu, Patrick Zhao, Motahhare Eslami, Jena D. Hwang, Faeze Brahman, Carolyn Rose, Maarten Sap · 4 December 2025

  Current LLMs are trained to refuse potentially harmful input queries regardless of whether users actually had harmful intents, causing a tradeoff between safety and user experience. Through a study of 480 participants evaluating 3,840 query-response pairs, we examine how different refusal strategies…

• A Guardrail for Safety Preservation: When Safety-Sensitive Subspace Meets Harmful-Resistant Null-Space

  Bingjie Zhang, Yibo Yang, Zhe Ren, Dandan Guo, Jindong Gu, Philip Torr, Bernard Ghanem · 28 October 2025

  Large language models (LLMs) have achieved remarkable success in diverse tasks, yet their safety alignment remains fragile during adaptation. Even when fine-tuning on benign data or with low-rank adaptation, pre-trained safety behaviors are easily degraded, leading to harmful responses in the fine-t…

• RSafe: Incentivizing proactive reasoning to build robust and adaptive LLM safeguards

  Jingnan Zheng, Xiangtian Ji, Yijun Lu, Chenhang Cui, Weixiang Zhao, Gelei Deng, Zhenkai Liang, An Zhang, Tat-Seng Chua · 27 October 2025