Machine learning, AI research, and applications.
Researchers developed APEX-Searcher, a system that improves how AI language models search for and use external knowledge to answer complex questions, which could make these models more useful for real-world tasks.
AI coding assistants can introduce bugs that break existing tests, but researchers propose a new "test-driven" approach to track code impact and reduce regressions.
AI safety systems that only check individual messages may miss risks that emerge over time, like concerning behavior patterns, posing system-level threats that could affect many people.
Researchers have developed a navigation system for autonomous wheelchairs that is designed to be socially aware, reducing the cognitive burden for users with mobility impairments and improving their ability to move around public spaces.
Researchers developed a more efficient way to translate between multiple languages using a new AI model, which could improve translation quality for less common languages.
OpenAI has agreed to give the US military access to its powerful AI technology, raising concerns about potential military applications and the ethical implications of advanced AI in warfare.
Researchers taught AI language models to generate code that uses private software libraries, enabling them to create more useful applications for developers. This advance could help make AI-generated code more practical and powerful for real-world software projects.
Researchers developed a new AI model that can accurately predict the behavior of light beams in complex environments, which could improve wireless communication technologies like 5G and beyond.
Researchers have developed a new AI system that can generate personalized content while preserving user privacy. This could lead to more customized services and apps that respect data privacy.
Researchers developed a new AI-powered modeling tool to help communities plan for climate challenges, which could make environmental planning more effective and inclusive for local populations.
Researchers found that using a single "anchor" model to judge the quality of other generative AI models is an efficient way to evaluate their performance, though it may not capture all nuances.
Autonomous AI agents are forming online communities on social platforms, raising new questions about how AI-driven social dynamics may impact human users. This research examines the emergent behaviors and implications of these AI-agent communities.
Researchers developed a new algorithm that can analyze 3D point cloud data more efficiently by incorporating rigid motion symmetries. This could improve 3D object detection and other applications relying on point cloud data.
Researchers developed a new machine learning technique that uses multi-dimensional sentiment analysis on news articles to better predict fluctuations in oil futures prices, which could help investors and consumers plan for volatile energy markets.
Researchers found ways to use "differential privacy" to secure sensitive data inside generative AI systems, allowing them to be used more safely in enterprise settings without compromising user privacy.
AI auto-complete can subtly sway people's views on social issues, as it suggests certain word choices over others when writing, potentially influencing how people think without their awareness.
Researchers developed a more efficient way to translate between multiple languages using a new AI model, which could improve translation quality for less common languages.
Researchers developed a new machine learning model that can more accurately measure the warping of light caused by distant galaxies, which helps us better understand dark matter and the structure of the universe.
Researchers developed a new technique to learn efficient policy representations without expensive expert demonstrations. This could lead to more accessible AI systems that can better imitate human decision-making.
A new AI-based tool called CRIMSON can help evaluate the quality of automatically generated medical reports, ensuring they are clinically accurate and relevant to patient care.
Researchers created a massive Pashto language dataset to help improve AI language models for the 60 million people who speak this underrepresented language.
A new AI system can automatically assess the quality of medical image translations, ensuring accuracy for critical applications like treatment planning where errors could impact patient outcomes.
Language models are now optimized to match human preferences, not just describe behavior - this could change how they are used for things like recommendation and decision-making.
Researchers developed a technique to improve how a state-of-the-art AI language model can adapt to new situations, which could lead to better real-world performance for AI assistants.
Researchers developed a system called TrajBooster that allows robots to quickly learn new manipulation tasks, even with limited training data. This could make humanoid robots more versatile and useful in real-world environments.
Researchers developed a new robot control method that can adapt to changing situations, improving safety and reliability of robots in complex environments. This could lead to more capable and responsive robots that can better handle real-world challenges.
Researchers have developed a way to cryptographically verify the model used in large language model (LLM) queries, preventing providers from substituting cheaper models or using cached responses. This could give users confidence that they are getting the expected AI inference.
A new AI system can automatically translate instruction manuals into robotic assembly plans that properly account for the physical constraints of connectors, enabling more reliable robotic assembly.
Researchers developed a new AI technique called PromptHub that can more effectively combine visual demonstrations to complete visual tasks, potentially improving real-world AI applications.