Latest Articles for Safety In Technology

Researchers develop new methods for training robots safely in risky environments.

Research highlights safety neurons' role in enhancing LLM safety and responsibility.

A new approach to enhance model safety through prediction rejection.

Research on magnetic islands enhances plasma stability and disruption prevention in tokamaks.

This article discusses methods for improving AI alignment with various cultures.

A new method helps identify weak points in deep learning models quickly.

Research reveals language models struggle with false reasoning, raising safety concerns.

Research focuses on managing plasma disruptions to improve fusion reactor safety.

CCL ensures neural networks maintain accuracy while learning new tasks.

A tool to analyze and improve computer image recognition errors.

UNRealNet enhances robot navigability in tough terrain using advanced techniques.

InferAct improves decision-making safety for AI agents in various tasks.

A new method improves robot walking safety and efficiency.

Researching how robots work together in shared spaces for safe interactions.

A new method to enhance safety in critical systems using language models.

This study analyzes the performance of neural network circuits and their reliability.

A new method improves understanding of safety constraints in robotics.

Examining how language models can refuse to answer for improved safety.

This article reviews how vector quantization impacts understanding decisions in reinforcement learning systems.

Learn how program verification ensures software reliability in critical industries.

A new method enhances RL agents' resilience against harmful input changes.

Improving fault detection and diagnostics in nuclear reactors using deep learning techniques.

This method improves safety in image generation while maintaining quality.

A new framework enhances robot safety and efficiency in unpredictable settings.

A new method improves safety in decision-making for machines.

A new approach to enhance how robots understand and respond to users.

LEVIS helps find safe input spaces for reliable neural network outputs.

Caution-Aware Transfer enhances safety and performance in reinforcement learning applications.

A new method improves detection of harmful prompts in language models.

This work focuses on explaining decision-making in AI using Monte Carlo Tree Search.

Introducing CBF-LLM: a method for safer text generation in LLMs.

A study on false refusals in language models and their impact on user experience.

A new method combines reinforcement learning and safety to enhance robot tasks.

A framework for ensuring robots perform safely and effectively in human interactions.

This article discusses ways to enhance safety in RL using language models.

A method to assess AI agents' evaluations for safety and reliability.

A new method enhances detection of unexpected data in machine learning models.

RADER system improves robotic learning through safe demonstrations in extended reality.

Examining how training data impacts language model outputs and safety measures.

New training method improves LLM safety and performance.