A smarter way for large language models to think about hard problems
To make large language models (LLMs) more accurate when answering harder questions, researchers can let the model spend more time thinking about potential solutions.
computer-sciencesPlatforms like TikTok could boost viewing times by grouping users to better match their preferences
In the nine years since TikTok debuted, it's helped transform the way people view and absorb information, along with other short-form video platforms such as Instagram and Snapchat. Every month, TikTok alone has nearly 1.6 billion active users globally.
computer-sciences새로운 아이디어가 현실로, 연구자 창업 문화 확산 선도
새로운 아이디어가 현실로, 연구자 창업 문화 확산 선도 – 2025 한국기계연구원 창업 아이디어 경진대회 개최 – [2025 창업 아이디어 경진대회 시상식 단체사진] □ 한국기계연구원(원장 류석현, 이하 기계연)은 12월 3일 (수) 대전 본원 대회의실에서 ‘2025 한국기계연구원 창업 아
한국기계연구원 > KIMM NEWS
Open-source framework enables addition of AI to software without prompt engineering
Developers can now integrate large language models directly into their existing software using a single line of code, with no manual prompt engineering required. The open-source framework, known as byLLM, automatically generates context-aware prompts based on the meaning and structure of the program, helping developers avoid hand-crafting detailed prompts, according to a conference paper presented at the SPLASH conference in Singapore in October 2025 and published in the Proceedings of the ACM on Programming Languages.
computer-sciencesGuided learning helps previously 'untrainable' neural networks learn effectively
Even networks long considered "untrainable" can learn effectively with a bit of a helping hand. Researchers at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) have shown that a brief period of alignment between neural networks, a method they call guidance, can dramatically improve the performance of architectures previously thought unsuitable for modern tasks.
computer-sciencesAI tool created to help sight-impaired programmers
A University of Texas at Dallas researcher and his collaborators have developed an artificial intelligence (AI)-assisted tool that makes it possible for visually impaired computer programmers to create, edit and verify 3D models independently.
computer-sciencesBig data in the AFL: Changing the game, one stat at a time
When Darren Jarman kicked an astounding five goals in the last quarter of the 1997 AFL Grand Final, I was 8 years old.
computer-sciencesNew control system teaches soft robots the art of staying safe
Imagine having a continuum soft robotic arm bend around a bunch of grapes or broccoli, adjusting its grip in real time as it lifts the object. Unlike traditional rigid robots that generally aim to avoid contact with the environment as much as possible and stay far away from humans for safety reasons, this arm senses subtle forces, stretching and flexing in ways that mimic more of the compliance of a human hand. Its every motion is calculated to avoid excessive force while achieving the task efficiently.
computer-sciencesApproximate domain unlearning: Enabling safer and more controllable vision-language models
Vision-language model (VLM) is a core technology of modern artificial intelligence (AI), and it can be used to represent different forms of expression or learning, such as photographs, illustrations, and sketches.
computer-sciencesEnhancing navigability for tributaries
Inland waterway transportation has played a limited role in Europe so far, with a share of about 6%. Together with 15 partners, Fraunhofer researchers are seeking to change this with the EU project CRISTAL.
computer-sciences
AI learns to 'listen': Compact speech tokens help models understand spoken words
Large language models (LLMs) such as ChatGPT and Gemini were originally designed to work with text only. Today, they have evolved into systems that can work with many types of information at once (multimodal systems), as well as understand and generate images, audio, speech and music.
computer-sciencesResearchers extend tensor programming to the continuous world
When the FORTRAN programming language debuted in 1957, it transformed how scientists and engineers programmed computers. Complex calculations could suddenly be expressed in concise, math-like notation using arrays—collections of values that make it easier to describe operations on data. That simple idea evolved into today's "tensors," which power many of the world's most advanced AI and scientific computing systems through modern frameworks like NumPy and PyTorch.
computer-sciencesHumans and AI models show similar confusion when reading tricky program code
Researchers from Saarland University and the Max Planck Institute for Software Systems have, for the first time, shown that the reactions of humans and large language models (LLMs) to complex or misleading program code significantly align, by comparing brain activity of study participants with model uncertainty.
computer-sciences
Visualizing the internal structure behind AI decision-making
Although deep learning–based image recognition technology is rapidly advancing, it still remains difficult to clearly explain the criteria AI uses internally to observe and judge images. In particular, technologies that analyze how large-scale models combine various concepts (e.g., cat ears, car wheels) to reach a conclusion have long been recognized as a major unsolved challenge.
computer-sciencesSix criteria for the reliability of AI
Language models based on artificial intelligence (AI) can answer any question, but not always correctly. It would be helpful for users to know how reliable an AI system is. A team at Ruhr University Bochum and TU Dortmund University suggests six dimensions that determine the trustworthiness of a system, regardless of whether the system is made up of individuals, institutions, conventional machines, or AI.
computer-sciencesAI decodes pianists' muscle activity via video
AI and human-movement research intersect in a study that enables precise estimation of hand muscle activity from standard video recordings. Using a deep-learning framework trained on a large, comprehensive multimodal dataset from professional pianists, the researchers introduce a system that accurately reconstructs muscle activation patterns without sensors.
computer-sciencesNew model measures how AI sycophancy affects chatbot accuracy and rationality
If you've spent any time with ChatGPT or another AI chatbot, you've probably noticed they are intensely, almost overbearingly, agreeable. They apologize, flatter and constantly change their "opinions" to fit yours.
computer-sciencesLLMs use grammar shortcuts that undermine reasoning, creating reliability risks
Large language models (LLMs) sometimes learn the wrong lessons, according to an MIT study. Rather than answering a query based on domain knowledge, an LLM could respond by leveraging grammatical patterns it learned during training. This can cause a model to fail unexpectedly when deployed on new tasks.
computer-sciences
