Recent technological advances are enabling the development of computational tools that could significantly improve the quality of life of individuals with disabilities or sensory impairments. These include so-called electromyography-to-speech (ETS) conversion models,...
Machine learning & AI
A method to mitigate hallucinations in large language models
Large language models (LLMs), artificial neural networks-based architectures that can process, generate and manipulate texts in various human languages, have recently become increasingly widespread. These models are now being used in a wide range of settings, to...
Tackling long-range malware detection tasks using holographic global convolutional networks
Over the past few decades, cyber-attackers have devised increasingly sophisticated malware that can disrupt the functioning of computer systems or grant them access to sensitive data. The development of techniques that can reliably detect the presence of malware and...
A framework to detect hallucinations in the text generated by LLMs
Large language models (LLMs) are advanced AI-based dialogue systems that can answer user queries and generate convincing texts following human instructions. After the advent of ChatGPT, the highly performing model developed by OpenAI, these models have become...
A framework to enhance the safety of text-to-image generation networks
The emergence of machine learning algorithms that can generate texts and images following human users' instructions has opened new possibilities for the low-cost creation of specific content. A class of these algorithms that are radically transforming creative...
Study explores why human-inspired machines can be perceived as eerie
Artificial intelligence (AI) algorithms and robots are becoming increasingly advanced, exhibiting capabilities that vaguely resemble those of humans. The growing similarities between AIs and humans could ultimately bring users to attribute human feelings, experiences,...
Emulating neurodegeneration and aging in artificial intelligence systems
In recent years, developers have introduced artificial intelligence (AI) systems that can simulate or reproduce various human abilities, such as recognizing objects in images, answering questions, and more. Yet in contrast with the human mind, which can deteriorate...
A new framework to generate human motions from language prompts
Machine learning-based models that can autonomously generate various types of content have become increasingly advanced over the past few years. These frameworks have opened new possibilities for filmmaking and for compiling datasets to train robotics algorithms.
A miniaturized vision-based tactile sensor based on fiber optic bundles
Researchers at Meta AI, Stanford University, Technische, Universität Dresden and the German Cancer Research Center (DFKZ) recently developed DIGIT Pinki, a miniature-sized sensor that can detect tactile information. This sensor, presented in a paper posted...
A scalable reinforcement learning–based framework to facilitate the teleoperation of humanoid robots
The effective operation of robots from a distance, also known as teleoperation, could allow humans to complete a vast range of manual tasks remotely, including risky and complex procedures. Yet teleoperation could also be used to compile datasets of human motions,...