New AI model accurately grades messy handwritten math answers and explains student errors
A research team affiliated with UNIST has unveiled a novel AI system capable of grading and providing detailed feedback on even the most untidy handwritten math answers—much like a human instructor.
computer-sciencesAI models stumble on basic multiplication without special training methods, study finds
These days, large language models can handle increasingly complex tasks, writing complex code and engaging in sophisticated reasoning. But when it comes to four-digit multiplication, a task taught in elementary school, even state-of-the-art systems fail. Why?
computer-sciencesFor computational devices, talk isn't cheap: Research reveals unavoidable energy costs across all communication channels
Every task we perform on a computer—whether number crunching, watching a video, or typing out an article—requires different components of the machine to interact with one another. "Communication is massively crucial for any computation," says former SFI Graduate Fellow Abhishek Yadav, a Ph.D. scholar at the University of New Mexico. But scientists don't fully grasp how much energy computational devices spend on communication.
computer-sciences
Multi-agent AI could change everything—if researchers can figure out the risks
You might have seen headlines sounding the alarm about the safety of an emerging technology called agentic AI.
computer-sciencesOne pull of a string is all it takes to deploy these complex structures
MIT researchers have developed a new method for designing 3D structures that can be transformed from a flat configuration into their curved, fully formed shape with only a single pull of a string.
computer-sciencesHelping AI agents search to get the best results out of large language models
Whether you're a scientist brainstorming research ideas or a CEO hoping to automate a task in human resources or finance, you'll find that artificial intelligence (AI) tools are becoming the assistants you didn't know you needed. In particular, many professionals are tapping into the talents of semi-autonomous software systems called AI agents, which can call on AI at specific points to solve problems and complete tasks.
computer-sciencesNew computer vision method links photos to floor plans with pixel-level accuracy
For people, matching what they see on the ground to a map is second nature. For computers, it has been a major challenge. A Cornell research team has introduced a new method that helps machines make these connections—an advance that could improve robotics, navigation systems, and 3D modeling.
computer-sciencesGenerative AIs fail at the game of visual 'telephone'
Generative AIs may not be as creative as we assume. Publishing in the journal Patterns, researchers show that when image-generating and image-describing AIs pass the same descriptive scene back and forth, they quickly veer off topic.
computer-sciencesFlexible position encoding helps LLMs follow complex instructions and shifting states
Most languages use word position and sentence structure to extract meaning. For example, "The cat sat on the box," is not the same as "The box was on the cat." Over a long text, like a financial document or a novel, the syntax of these words likely evolves.
computer-sciencesAI gets a private tutor for learning human preferences more accurately
No matter how much data they learn, why do artificial intelligence (AI) models often miss the mark on human intent? Conventional comparison learning, designed to help AI understand human preferences, has frequently led to confusion rather than clarity. A KAIST research team has now presented a new learning solution that allows AI to accurately learn human preferences even with limited data by assigning it a "private tutor."
computer-sciences
New 3D benchmark leaves AI in knots
Today's artificial intelligence models can't even tie their own shoes.
computer-sciencesHow AI helps solve problems it doesn't even understand
Researchers at TU Wien have discovered an unexpected connection between two very different areas of artificial intelligence: Large Language Models (LLMs) can help solve logical problems—without actually "understanding" them.
computer-sciencesResolving to spend less time on your smartphone? Understanding your travel habits can help, say researchers
If you open a banking app, play a mobile game or scroll through a news feed every day while riding the bus, your commuting routine is probably bolstering your smartphone habit, according to new research that shows phone tendencies are stronger in locations chosen automatically.
computer-sciencesResearchers reveal bias in a widely used measure of algorithm performance
When scientists test algorithms that sort or classify data, they often turn to a trusted tool called Normalized Mutual Information (or NMI) to measure how well an algorithm's output matches reality. But according to new research, that tool may not be as reliable as many assume.
computer-sciencesEnabling small language models to solve complex reasoning tasks
As language models (LMs) improve at tasks like image generation, trivia questions, and simple math, you might think that human-like reasoning is around the corner. In reality, they still trail us by a wide margin on complex tasks. Try playing Sudoku with one, for instance, where you fill in numbers one through nine in such a way that each appears only once across the columns, rows, and sections of a nine-by-nine grid. Your AI opponent will either fail to fill in boxes on its own or do so inefficiently, although it can verify if you've filled yours out correctly.
computer-sciencesAI agents debate their way to improved mathematical reasoning
Large language models (LLMs), artificial intelligence (AI) systems that can process and generate texts in various languages, are now widely used worldwide to create written content, source information and even to code websites or applications. While these models have improved significantly over the past couple of years, their answers can sometimes contain factual inaccuracies or logical inconsistencies.
computer-sciencesNew system efficiently explains AI judgments in real-time
A research team led by Professor Jaesik Choi of KAIST's Kim Jaechul Graduate School of AI, in collaboration with KakaoBank Corp, has developed an accelerated explanation technology that can explain the basis of an artificial intelligence (AI) model's judgment in real-time.
computer-sciences'Periodic table' for AI methods aims to drive innovation
Artificial intelligence is increasingly used to integrate and analyze multiple types of data formats, such as text, images, audio and video. One challenge slowing advances in multimodal AI, however, is the process of choosing the algorithmic method best aligned to the specific task an AI system needs to perform.
computer-sciences
