An AI model builder developed at UC Irvine can autonomously design particle physics theories, a task traditionally only carried out by humans. The approach enables efficient exploration of new theoretical frameworks, helping uncover promising neutrino models for scientific study. Applications could expand to other theoretical model-building problems in the future. The National Science Foundation and the Department of Energy provided research support.

Researchers designed a robot that can swim underwater, then flap out of the water to continue flying through air, much like diving birds.

Researchers led by Kexin Huang report the development of a general-purpose biomedical artificial intelligence agent that can autonomously perform a range of research tasks. The authors tested their tool in various case studies, where it showed strong performance, and they say their results point “toward a future in which AI agents work alongside human researchers to accelerate biomedical discovery from basic research to translation.” Biomedical research drives discoveries in disease mechanisms, diagnostics, and therapeutics, but increasingly complex workflows, expanding datasets, specialized tools, and a rapidly growing literature make it difficult to fully leverage available knowledge and resources. As a result, valuable data and analytical opportunities often go underused because the demand for expert researchers exceeds the available workforce. This underscores the need for approaches that can scale expertise and accelerate discovery. The development of AI agents offers opportunities to reshape aspects of biomedical research, particularly if such tools could handle a broad range of biomedical tasks.

 

Here, Huang and colleagues introduce Biomni, an AI agent that mines biomedical literature to define comprehensive “action spaces.” The large language model also acts as the planning engine, interpreting a user request and preparing a multi-step series of sub-tasks to satisfy that request. In doing so, Biomni can compose and execute workflows without pre-set templates. The authors tested Biomni against a variety of biomedical benchmarks. They also evaluated how it performed compared to humans across several tasks, finding Biomni consistently matched expert accuracy while requiring substantially less time. They also tested the tool in five case studies ranging from analyzing wearable sensor data to designing laboratory protocols to assist wet-lab researchers. The authors note that while Biomni’s environment spans a wide range of biomedical tools and databases, “the evaluated tasks represent only a subset of the field, and key domains remain unexplored.” Even so, they say, by automating complex workflows, “Biomni enables researchers to redirect their efforts toward creative hypothesis generation, experimental innovation, and cross-disciplinary collaboration.”

 

***Related embargoed news briefing is planned for 11:00 a.m. U.S. ET on Tuesday, July 7, as a Zoom Webinar. Reporters interested in joining the briefing should please pre-register here https://aaas.zoom.us/webinar/register/WN_yEqbHT68TvqMItVMwEGOKg as soon as possible. See Other Information for Journalists section for full briefing information.***

A new statistical tool could help researchers and practitioners determine when personalized treatments, educational programs, or recommendations are truly worth the extra cost and complexity, according to a new study. The findings offer a rigorous tool to identify when tailoring interventions is likely to outperform more conventional one-size-fits-all approaches. Personalized interventions are increasingly used in fields such as medicine, education, marketing, economics, and computer science because individuals often respond differently to the same treatment or support. However, personalized interventions are more expensive, often require more data, and can be more difficult to implement than universal interventions. There are very few tools to evaluate and identify scenarios when the added cost and complexity of personalization are justified by meaningful improvements in outcomes. To address this, Zhaoqi Li and Emma Brunskill developed a statistical hypothesis test – the K-fold Personalization Test (KPT) – that analyzes existing datasets to evaluate whether a personalized intervention policy is expected to outperform the best universal intervention. Unlike previous methods, the test accommodates multiple interventions, many individual characteristics, and flexible machine learning models while maintaining rigorous statistical guarantees, including control of false-positive (Type I) error. Across four datasets spanning job training, depression treatment, education, and marketing systems, the authors found that the KPT approach consistently demonstrated broad applicability and outperformed previous statistical methods for evaluating personalization. Li and Brunskill note, however, that the test evaluates personalization only within a user-specified class of decision policies and does not identify the optimal personalized policy itself. Yet even with these limitations, the KPT provides a rigorous and broadly applicable framework for helping researchers and practitioners determine when personalization is likely to deliver meaningful benefits.

A flapping-wing robot that can both swim underwater and fly through the air is helping scientists rethink how diving birds manage life in two radically different worlds. “In addition to shedding light on the morphological and behavioral adaptations of aerial-aquatic animals, the design principles described here lay the foundation for a class of robots that can be used for limnology, oceanography, marine ecosystem monitoring, and coastal management,” the authors write. Roughly 100 bird species are capable of both flying through the air and propelling themselves underwater using only their wings. While these birds use a similar flapping motion in both environments, they adjust by slowing their wingbeats and reducing their wing area underwater. Because water is nearly 1,000 times denser than air, moving efficiently in each medium requires very different forces and wing movements. However, studying these behaviors and movements in live animals is challenging, and computer simulations struggle to model the complex interactions between flapping wings, fluid forces, and the transition from water to air accurately. As a result, the physiological adaptations and compromises that allow birds to move through such disparate environments efficiently remain relatively unknown. According to Raphael Zufferey and colleagues, robotic models offer a valuable alternative because they adhere the same physical principles as living animals while enabling researchers to control their design and movement precisely.

 

In this study, Zufferey et al. present flapping-wing robots capable of flying through the air, swimming underwater, and transitioning seamlessly between the two environments – aerial-aquatic vehicles designed to explore the physical challenges faced by diving birds and the design strategies that make dual-environment locomotion possible. The modular, 250-gram robot features a streamlined fuselage, two flexible membrane wings, and a movable tail. It is also fully waterproof, untethered, and equipped with onboard electronics, allowing the authors to adjust wing-flapping frequency and tail position wirelessly, to examine systematically how wing size, flexibility, and flapping influence movement in air, underwater, and during transitions between the two environments. By comparing data from diving birds with their experimental observations, Zufferey et al. found that complex wing-folding mechanisms are not essential for aerial-aquatic locomotion. Instead, an effective balance of wing flexibility, size, and flapping frequency is sufficient to achieve similar performance. What’s more, experiments revealed that smaller wings increase underwater speed but do not improve swimming efficiency, suggesting that reduced wing size in diving birds may primarily enhance maneuverability and prey pursuit rather than conserve energy. Wings with intermediate flexibility provided the best overall performance, improving underwater propulsion while still generating sufficient lift for flight. Because flying requires less energy than swimming, the authors discovered that it was more efficient for the robot to leave the water and fly than to remain submerged over longer distances. The study also showed that wing-powered takeoff from the water is possible without leg assistance, although it requires substantial power.

In experiments in which physicians made decisions about treating hypothetical patients, the physicians tended to trust incorrect advice presented as being generated by artificial intelligence (AI), even after given the opportunity to notice that patient recovery data contradicted the recommendations. Aranzazu Vinas of the University of the Basque Country, Spain, and colleagues present these findings in the open-access journal PLOS Digital Health. 

In creating a comprehensive, AI-enabled research agent for the biomedical sciences, Stanford researchers hope to speed innovation by eliminating the tedium of scientific legwork.