Social media, such as Twitter, provides opportunities for caregivers of dementia patients to share their experiences and seek support for a variety of reasons. Availability of this information online also paves the way for the development of internet-based interventions in their support. However, for this purpose, tweets written by caregivers of dementia patients must first be identified. This paper demonstrates our system for the SMM4H 2025 shared task 3, which focuses on detecting tweets posted by individuals who have a family member with dementia. The task is outlined as a binary classification problem, differentiating between tweets that mention dementia in the context of a family member and those that do not. Our solution to this problem explores large language models (LLMs) with various prompting methods. Our results show that a simple zero-shot prompt on a fine-tuned model yielded the best results. Our final system achieved a macro F1-score of 0.95 on the validation set and the test set. Our full code is available on GitHub.

The report describes the discussions from the Workshop on Mathematical Opportunities in Digital Twins (MATH-DT) from December 11-13, 2023, George Mason University. It illustrates that foundational Mathematical advances are required for Digital Twins (DTs) that are different from traditional approaches. A traditional model, in biology, physics, engineering or medicine, starts with a generic physical law (e.g., equations) and is often a simplification of reality. A DT starts with a specific ecosystem, object or person (e.g., personalized care) representing reality, requiring multi -scale, -physics modeling and coupling. Thus, these processes begin at opposite ends of the simulation and modeling pipeline, requiring different reliability criteria and uncertainty assessments. Additionally, unlike existing approaches, a DT assists humans to make decisions for the physical system, which (via sensors) in turn feeds data into the DT, and operates for the life of the physical system. While some of the foundational mathematical research can be done without a specific application context, one must also keep specific applications in mind for DTs. E.g., modeling a bridge or a biological system (a patient), or a socio-technical system (a city) is very different. The models range from differential equations (deterministic/uncertain) in engineering, to stochastic in biology, including agent-based. These are multi-scale hybrid models or large scale (multi-objective) optimization problems under uncertainty. There are no universal models or approaches. For e.g., Kalman filters for forecasting might work in engineering, but can fail in biomedical domain. Ad hoc studies, with limited systematic work, have shown that AI/ML methods can fail for simple engineering systems and can work well for biomedical problems. A list of `Mathematical Opportunities and Challenges' concludes the report.

This paper describes the full end-to-end design of our primary scoring agent in an aerial autonomous robotics competition from April 2023. As open-ended robotics competitions become more popular, we wish to begin documenting successful team designs and approaches. The intended audience of this paper is not only any future or potential participant in this particular national Defend The Republic (DTR) competition, but rather anyone thinking about designing their first robot or system to be entered in a competition with clear goals. Future DTR participants can and should either build on the ideas here, or find new alternate strategies that can defeat the most successful design last time. For non-DTR participants but students interested in robotics competitions, identifying the minimum viable system needed to be competitive is still important in helping manage time and prioritizing tasks that are crucial to competition success first.

Bad news for those who struggle with anxiety, get too competitive, or simply choke under pressure: new research suggests that you may have to stay at home on Earth while other, more laid back and'agreeable' types colonize Mars. The new study, which is still undergoing peer review, ran computer simulations tracking the progress of human settlements on the Red Planet through their first 28 years of virtual operation. 'Agreeable personality types were assessed to be the most enduring for the long term,' the researchers found, across all four of the personality types used in their simulations, 'whereas neurotics showed least adaptation capacity.' The researchers also discovered that the minimum number of settlers needed to successfully operate a human colony on Mars was much lower than previously expected: just 22 people. 'Contrary to other literature,' they wrote of their simulated Martian colonies, 'the minimum number of people with all personality types that can lead to a sustainable settlement is in the tens and not hundreds.'

Over at "SkepticalScience", which is neither skeptical nor scientific, they're hyping a new "Artificial Intelligence" (AI) tool developed by John Cook et al. to identify "denialist claims". The paper laying out this foolishness is in Nature Scientific Reports in an article with the most sciency title of "Computer-assisted classification of contrarian claims about climate change". "Ultimately, our goal is the Holy Grail of fact-checking, which is being able to detect and debunk misinformation in real time," said Cook, who partly developed the framework previously at George Mason University. Because in total contradiction to point 4 immediately above, that experts are not unreliable, one of the finest physicists of my lifetime, Richard Feynman, famously said: Nature Magazine, a premier scientific journal and a huge defender of the anthropogenic climate change hypothesis, has an article on the subject which says: So clearly, Nature Magazine is a secret nest of climate "denialists" whose claims should be censored before anyone can be misled by them … and while that example alone should be enough to totally discredit their artificial stupidity, it's just the first of many. So it's gonna identify articles pointing out that while in most of the media heatwaves are always explained as climate change, cold spells are just plain old weather … For most species, including humans and coral reefs, a change of a degree in average temperature over fifty years means nothing.

Much scientific enquiry across disciplines is founded upon a mechanistic treatment of dynamic systems that ties form to function. A highly visible instance of this is in molecular biology, where an important goal is to determine functionally-relevant forms/structures that a protein molecule employs to interact with molecular partners in the living cell. This goal is typically pursued under the umbrella of stochastic optimization with algorithms that optimize a scoring function. Research repeatedly shows that current scoring function, though steadily improving, correlate weakly with molecular activity. Inspired by recent momentum in generative deep learning, this paper proposes and evaluates an alternative approach to generating functionally-relevant three-dimensional structures of a protein. Though typically deep generative models struggle with highly-structured data, the work presented here circumvents this challenge via graph-generative models. A comprehensive evaluation of several deep architectures shows the promise of generative models in directly revealing the latent space for sampling novel tertiary structures, as well as in highlighting axes/factors that carry structural meaning and open the black box often associated with deep models. The work presented here is a first step towards interpretative, deep generative models becoming viable and informative complementary approaches to protein structure prediction.

Delivery robots incorporating AI are on the march, being deployed more widely on the ground, sometimes crowding sidewalks. Delivery robot providers include Starship Technologies, a startup created by Janus Friis and Ahti Heinla, founders of Skype. The company offers a general-purpose home delivery robot that today is an array of cameras and GPS sensors, but in the future will include microphones, speaker, and the ability via AI-driven natural language processing, to talk to customers. Since 2016, Starship has carried out 50,000 delivers in over 100 cities across 20 countries, according to an account in SingularityHub written by Dr. Peter H. Diamandis, the founder of Singularity University and the founder and executive chairman of the XPrize Foundation. Another startup delivery provider is Nuro, co-founded by Jiajun Zhu, an engineer who helped develop Google's self-driving car.

MINNEAPOLIS--(BUSINESS WIRE)-- U.S. Bank (USBK) has hired technology leader Dr. Tanushree Luke to lead Artificial Intelligence (AI) efforts at the company. In this role, she will drive the continued development of the AI practice within the U.S. Bank Innovation group and AI strategies across the enterprise. This press release features multimedia. U.S. Bank has hired technology leader Dr. Tanushree Luke to lead Artificial Intelligence (AI) efforts at the company. Dr. Luke's career has spanned multiple industries and sectors.

Bank has hired technology leader Dr. Tanushree Luke to lead Artificial Intelligence (AI) efforts at the company. In this role, she will drive the continued development of the AI practice within the U.S. Bank Innovation group and AI strategies across the enterprise. Dr. Luke's career has spanned multiple industries and sectors. She joins U.S. Bank from Capital One, where she served as the Head of Predictive AI and Machine Learning for Capital One's Conversational AI Platform/Eno. Prior to that, Dr. Luke was the Chief Data Scientist at BitVoyant and was also previously a Technical Lead for the Department of Defense/DARPA Network Defense Program.

At George Mason University in Virginia, a fleet of several dozen autonomous robots deliver food to students on campus. At George Mason University in Virginia, a fleet of several dozen autonomous robots deliver food to students on campus. George Mason University looks like any other big college campus with its tall buildings, student housing, and manicured green lawns – except for the robots. This Northern Virginia university recently set up several dozen meal delivery robots from Starship Technologies to make it easier for students to access food. Multiple colleges across the country have deployed delivery robots – including University of the Pacific in Stockton, Calif., and Northern Arizona University – but George Mason University is the first college in the United States to incorporate robots into its student dining plan. The school is partnering with food service provider Sodexo for the program.