Individual fairness
Individual fairness is a concept in ethics and machine learning that emphasizes the fair treatment of individuals by ensuring that similar individuals receive similar outcomes.
Technology & Innovation
Jailbreak (large language models)
Jailbreak in large language models refers to techniques used to bypass or override the built-in restrictions and safety mechanisms of AI systems to produce outputs that are otherwise restricted. These methods raise ethical, security, and safety concerns within the AI community.
Avatar representation learning
Avatar representation learning involves the development of computational models that capture and encode the characteristics and behaviors of digital avatars. It aims to create more realistic, adaptive, and personalized avatars through machine learning techniques.
Field-programmable gate array for AI
Field-programmable gate arrays (FPGAs) are integrated circuits that can be configured post-manufacturing to perform specialized tasks. In the context of artificial intelligence (AI), FPGAs provide customizable hardware acceleration, balancing flexibility and performance for AI workloads.
Computer vision
Computer vision is a multidisciplinary field that enables computers to interpret and process visual information from the world. It involves the development of algorithms and systems that can analyze images and videos to extract meaningful data.
XLM (Cross-lingual Language Model)
XLM (Cross-lingual Language Model) is a type of pretrained neural network model designed to understand and generate text across multiple languages. It leverages multilingual training data to improve performance on various natural language processing tasks involving different languages.
Proximal policy optimization (PPO) for language models
Proximal policy optimization (PPO) is a reinforcement learning algorithm adapted for training language models by optimizing policies that generate text. It enables efficient and stable fine-tuning of language models with human feedback or reward signals.
Causal reinforcement learning
Causal reinforcement learning integrates causal inference with reinforcement learning, enabling agents to make decisions based on cause-and-effect relationships.
Shape-E
Shape-E is an advanced robotic platform designed for research and development in autonomous navigation and manipulation. It integrates multiple sensors and actuators to perform complex tasks in dynamic environments, often used in academic and industrial contexts.
SMPL (skinned multi-person linear model)
The Skinned Multi-Person Linear (SMPL) model is a widely used parametric human body model that captures the shape and pose of individuals in a 3D space.