Latest Articles for Safety

A new method for safely controlling unpredictable systems.

Examining vulnerabilities and safety strategies for LLM-powered scientific agents.

A study examines the fragile safety mechanisms in language models and proposes improvements.

Researchers use B-splines to create smoother, safer paths for autonomous racing.

A new method enhances how robots interpret user instructions safely.

Research highlights vaccination benefits for cancer patients with weakened immune systems.

Exploring the potential of eVTOL vehicles in managing urban transportation challenges.

Examining the traits and challenges of developing intelligent machines.

A new method aims to reduce harmful outputs from AI language models.

Examining the benefits and risks of modern sensor technology.

Research improves aircraft ditching load predictions using machine learning techniques.

Examining how AVs can improve road safety while maintaining traffic flow.

Introducing PREDILECT, a method for efficient human-robot interaction.

This paper examines invariant checking in systems with infinite states using two algorithms.

A wearable system helps workers lift safely and avoid injuries.

Study reveals how drivers prefer communication while distracted.

A novel method improves robot learning using 3D images.

AI companies must support independent research for safer systems.

Introducing a new language for clear and efficient access permissions.

Insights into parental concerns about children's VR safety and privacy.

A new method enhances communication skills of language agents.

A new method enhances safety features in multimodal AI systems without extensive training.

Exploring the key traits and challenges of developing reliable AI systems.

Gillian-Rust offers a novel way to verify unsafe Rust code effectively.

A study on the performance and evaluation methods of large language models.

Two new algorithms enhance robot perception and ensure safety in various environments.

An overview of formal methods for designing safe autonomous systems.

New techniques enable safer decision-making in reinforcement learning through human expert guidance.

ALERT benchmark assesses safety risks in language models to improve their responses.

A review of datasets focused on enhancing LLM safety.

Cendol enhances language tech for Indonesia's diverse languages and cultures.

An overview of creating effective reward functions in self-driving cars.

New control method offers a realistic approach for improved system safety and efficiency.

New framework aims to enhance drug dosage recommendations during pregnancy.

New techniques aim to maintain safety in unpredictable control systems with limited data.

Measuring robustness and safety in cyber-physical systems under sensor threats.

Robotic systems are shaping the future of neurosurgery, improving safety and precision.

Study reveals how models respond to gibberish prompts and implications for safety.

Exploring the responsible use of generative AI technology in various fields.

A new approach to safety analysis using piecewise stochastic barrier functions.