Clinical Trials

Rethinking Bottlenecks in Safety Fine-Tuning of Vision Language Models

Large Vision-Language Models (VLMs) have achieved remarkable performance across a wide range of tasks. However, their deployment in safety-critical domains poses significant challenges. Existing safety fine-tuning methods, which focus on textual or multimodal content, fall short in addressing challenging cases or disrupt the balance between helpfulness and harmlessness. Our evaluation highlights...

"My Whereabouts, my Location, it's Directly Linked to my Physical Security": An Exploratory Qualitative Study of Location-Dependent Security and Privacy Perceptions among Activist Tech Users

Digital-safety research with at-risk users is particularly urgent. At-risk users are more likely to be digitally attacked or targeted by surveillance and could be disproportionately harmed by attacks that facilitate physical assaults. One group of such at-risk users are activists and politically active individuals. For them, as for other at-risk users, the rise of smart environments harbors new ...

Towards Safe AI Clinicians: A Comprehensive Study on Large Language Model Jailbreaking in Healthcare

Large language models (LLMs) are increasingly utilized in healthcare applications. However, their deployment in clinical practice raises significant...

Beyond Benchmarks: On The False Promise of AI Regulation

The rapid advancement of artificial intelligence (AI) systems in critical domains like healthcare, justice, and social services has sparked numerous...

Force-Based Robotic Imitation Learning: A Two-Phase Approach for Construction Assembly Tasks

The drive for efficiency and safety in construction has boosted the role of robotics and automation. However, complex tasks like welding and pipe in...

An Empirical Study on LLM-based Classification of Requirements-related Provisions in Food-safety Regulations

As Industry 4.0 transforms the food industry, the role of software in achieving compliance with food-safety regulations is becoming increasingly cri...

Fat-to-Thin Policy Optimization: Offline RL with Sparse Policies

Sparse continuous policies are distributions that can choose some actions at random yet keep strictly zero probability for the other actions, which ...

Enhanced PEC-YOLO for Detecting Improper Safety Gear Wearing Among Power Line Workers

To address the high risks associated with improper use of safety gear in complex power line environments, where target occlusion and large variance ...

Provably-Safe Neural Network Training Using Hybrid Zonotope Reachability Analysis

Even though neural networks are being increasingly deployed in safety-critical control applications, it remains difficult to enforce constraints on ...

T2ISafety: Benchmark for Assessing Fairness, Toxicity, and Privacy in Image Generation

Text-to-image (T2I) models have rapidly advanced, enabling the generation of high-quality images from text prompts across various domains. However, ...

The Dual-use Dilemma in LLMs: Do Empowering Ethical Capacities Make a Degraded Utility?

Recent years have witnessed extensive efforts to enhance Large Language Models (LLMs) across various domains, alongside growing attention to their e...

Development of Application-Specific Large Language Models to Facilitate Research Ethics Review

Institutional review boards (IRBs) play a crucial role in ensuring the ethical conduct of human subjects research, but face challenges including inc...

Latent-space adversarial training with post-aware calibration for defending large language models against jailbreak attacks

Ensuring safety alignment has become a critical requirement for large language models (LLMs), particularly given their widespread deployment in real...

MSTS: A Multimodal Safety Test Suite for Vision-Language Models

Vision-language models (VLMs), which process image and text inputs, are increasingly integrated into chat assistants and other consumer AI applicati...

A Vessel Bifurcation Landmark Pair Dataset for Abdominal CT Deformable Image Registration (DIR) Validation

Deformable image registration (DIR) is an enabling technology in many diagnostic and therapeutic tasks. Despite this, DIR algorithms have limited cl...

Bootstrapping Corner Cases: High-Resolution Inpainting for Safety Critical Detect and Avoid for Automated Flying

Modern machine learning techniques have shown tremendous potential, especially for object detection on camera images. For this reason, they are also...

Phase of Flight Classification in Aviation Safety using LSTM, GRU, and BiLSTM: A Case Study with ASN Dataset

Safety is the main concern in the aviation industry, where even minor operational issues can lead to serious consequences. This study addresses the ...

Aviation Safety Enhancement via NLP & Deep Learning: Classifying Flight Phases in ATSB Safety Reports

Aviation safety is paramount, demanding precise analysis of safety occurrences during different flight phases. This study employs Natural Language P...

Determining Disturbance Recovery Conditions by Inverse Sensitivity Minimization

Power systems naturally experience disturbances, some of which can damage equipment and disrupt consumers. It is important to quickly assess the lik...

Driver Age and Its Effect on Key Driving Metrics: Insights from Dynamic Vehicle Data

By 2030, the senior population aged 65 and older is expected to increase by over 50%, significantly raising the number of older drivers on the road....