Clinical Trials

Personalized Safety in LLMs: A Benchmark and A Planning-Based Agent Approach

Large language models (LLMs) typically generate identical or similar responses for all users given the same prompt, posing serious safety risks in high-stakes applications where user vulnerabilities differ widely. Existing safety evaluations primarily rely on context-independent metrics - such as factuality, bias, or toxicity - overlooking the fact that the same response may carry divergent risk...

SPIRAL integration of generative AI in an undergraduate creative media course: effects on self-efficacy and career outcome expectations

Computing education and computing students are rapidly integrating generative AI, but we know relatively little about how different pedagogical strategies for intentionally integrating generative AI affect students' self-efficacy and career interests. This study investigates a SPIRAL integration of generative AI (Skills Practiced Independently, Revisited with AI Later), implemented in an introdu...

CONCORD: Concept-Informed Diffusion for Dataset Distillation

Dataset distillation (DD) has witnessed significant progress in creating small datasets that encapsulate rich information from large original ones. ...

Understanding and Mitigating Overrefusal in LLMs from an Unveiling Perspective of Safety Decision Boundary

Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of tasks, yet they often refuse to answer legitimate quer...

Automating Safety Enhancement for LLM-based Agents with Synthetic Risk Scenarios

Large Language Model (LLM)-based agents are increasingly deployed in real-world applications such as "digital assistants, autonomous customer servic...

Chain-of-Lure: A Synthetic Narrative-Driven Approach to Compromise Large Language Models

In the era of rapid generative AI development, interactions between humans and large language models face significant misusing risks. Previous resea...

Towards Evaluating Proactive Risk Awareness of Multimodal Language Models

Human safety awareness gaps often prevent the timely recognition of everyday risks. In solving this problem, a proactive safety artificial intellige...

Alignment and Safety of Diffusion Models via Reinforcement Learning and Reward Modeling: A Survey

Diffusion models have emerged as leading generative models for images and other modalities, but aligning their outputs with human preferences and sa...

Safety and User Experience of a Generative Artificial Intelligence Digital Mental Health Intervention: Exploratory Randomized Controlled Trial.

BACKGROUND: General awareness and exposure to generative artificial intelligence (AI) have increased recently. This transformative technology has the ...

Deconfounded Warm-Start Thompson Sampling with Applications to Precision Medicine

Randomized clinical trials often require large patient cohorts before drawing definitive conclusions, yet abundant observational data from parallel ...

Robustifying Vision-Language Models via Dynamic Token Reweighting

Large vision-language models (VLMs) are highly vulnerable to jailbreak attacks that exploit visual-textual interactions to bypass safety guardrails....

From Evaluation to Defense: Advancing Safety in Video Large Language Models

While the safety risks of image-based large language models have been extensively studied, their video-based counterparts (Video LLMs) remain critic...

Offline Guarded Safe Reinforcement Learning for Medical Treatment Optimization Strategies

When applying offline reinforcement learning (RL) in healthcare scenarios, the out-of-distribution (OOD) issues pose significant risks, as inappropr...

Clinical Efficacy of Multimodal Exercise Telerehabilitation Based on AI for Chronic Nonspecific Low Back Pain: Randomized Controlled Trial.

BACKGROUND: Exercise therapy is strongly recommended as a treatment for chronic nonspecific low back pain (CNSLBP). However, therapist-guided exercise...

Increasing the ethnic diversity of senior leadership within the English National Health Service: using an artificial intelligence approach to evaluate inclusive recruitment strategies in hospital settings.

BACKGROUND: The English National Health Service (NHS) strives for a fair, diverse, and inclusive workplace, but Black and Minority Ethnic (BME) repres...

Real-Time Stress Monitoring, Detection, and Management in College Students: A Wearable Technology and Machine-Learning Approach

College students are increasingly affected by stress, anxiety, and depression, yet face barriers to traditional mental health care. This study evalu...

UniErase: Unlearning Token as a Universal Erasure Primitive for Language Models

Large language models require iterative updates to address challenges such as knowledge conflicts and outdated information (e.g., incorrect, private...

Are Vision-Language Models Safe in the Wild? A Meme-Based Benchmark Study

Rapid deployment of vision-language models (VLMs) magnifies safety risks, yet most evaluations rely on artificial images. This study asks: How safe ...

Better Safe Than Sorry? Overreaction Problem of Vision Language Models in Visual Emergency Recognition

Vision-Language Models (VLMs) have demonstrated impressive capabilities in understanding visual content, but their reliability in safety-critical co...

The Pursuit of Empathy: Evaluating Small Language Models for PTSD Dialogue Support

Can small language models with 0.5B to 5B parameters meaningfully engage in trauma-informed, empathetic dialogue for individuals with PTSD? We addre...