Researchers at Binghamton University have applied a 70-year-old theory of information to the viral word game Wordle, revealing how a carefully chosen first guess can dramatically improve a playerβs chances of solving the puzzle.
The post Scientists Use Math to Solve Viral Word Game βWordleβ with 99% Success Rate appeared first on Sci.News: Breaking Science News.
Physicists recently wondered what would happen if you tried to split a photon βͺββ¬ and they found some unexpected behavior that may transform the way we think about particles.
The experiment, in which researchers simulated a photon being sliced by a shutter under various conditions, showed that a severed photon can lead to a complex mixture of zero to infinitely-many photons β raising some big questions about the nature of particle interactions.
Photons are elementary packets of light, which means they are not made up of anything else. So what does it even mean to try splitting a photon? It may be possible thanks to wave-particle duality β a core principle of quantum mechanics, the bizarre physics of the very small.
According to wave-particle duality, a photon is not only a particle but also a wave. Using theoretical calculations, the researchers investigated what would happen if you sent this photon through a shutter and closed it while the photon was passing through, effectively cutting off the tail end of the photon wave.
"I think that most physicists would expect there to be a certain probability that you have zero photons and a certain probability that you have a single photon left after you have done this," Johannes Skaar, co-author of the new study and a professor of theoretical physics at the University of Oslo, told Live Science. "And that is approximately true, but it is not exactly true."
This brings up another strange aspect of quantum mechanics: its probabilistic nature. Particles exist as a cloud of probabilities stretching to infinity. Until a particle is observed, its properties, such as its position or energy, are in a superposition of possible values; all we can know are the chances of finding it in a certain state.
Through their calculations, Skaar and his colleagues determined how cutting a photon affects these probabilities. In their study, recently accepted in the journal Physical Review Letters, they found that it would create a complex mixture of photon states, including one with an infinite number of photons.
Each of these states has a probability that depends on how quickly the shutter cuts the photon. The expected number of photons becomes infinite only if the shutter is closed infinitely quickly. For realistic shutter speeds, even a thousand photons would be extremely unlikely.
This may sound very strange, but the quantum physicists were unfazed. In fact, what surprised Skaar and his colleagues was what happens if you make measurements of the cut photon from different perspectives.
"When you measure from one side of the shutter, then it will look like a single photon state," Skaar said. "Then, on the other side, it will look like a vacuum state β that means no photons. And that is very strange because the actual state globally is this mixture from zero to infinity."
The fact that these complex mixtures can be treated locally as very simple states raises fundamental questions about the nature of particles. Skaar said they are still reckoning with the full extent of these implications and they are now considering how this process could play out for other quantum particles, such as electrons.
They hope that by following this theoretical thread through, they may be able to develop a neater way of describing particle interactions. Currently, particles' infinite stretching means they have been interacting for an infinite amount of time. This then poses a problem for causality βͺββ¬ the order of cause and effect βͺββ¬ in particle interactions, the team said.
These new theoretical photons with a cutoff tail would not have this problem, meaning the causal link in an interaction would be clear, Skaar said. He admitted that there is much more work to be done to develop the theoretical description of this interaction. However, the new result is an important step toward describing particle interactions with a clear causal relationship, which Skaar described as the team's "ultimate goal."
Systematic Reviews and Meta-Analyses Face a Data Transparency Crisis Impacting Healthcare and Policy
Systematic reviews and meta-analyses occupy a cornerstone role in modern healthcare research by aggregating data from multiple studies to distill the most robust evidence possible. Their findings inform clinical guidelines, shape public health policies, and direct future research priorities. Yet, despite their importance, a troubling crisis in data transparency threatens to erode the reliability of these comprehensive syntheses. Recent research from Karolinska Institutet sheds new light on this issue, revealing that limited access to underlying raw data severely undermines the trustworthiness of meta-analytical conclusions, with potentially profound implications for healthcare decision-making.
Meta-analyses are designed to combine data across various independent studies, increasing statistical power and improving estimates of treatment effects or associations within populations. Ideally, such analyses rely on access to raw participant-level data, which allows for complex and nuanced re-analyses that can account for heterogeneity, detect biases, and evaluate assumptions. However, researchers frequently find themselves constrained by incomplete, aggregate-level summaries published in original study reports. This forces analysts to perform calculations and imputations on limited detail, often based on unverified assumptions, which can skew the meta-analytic results in subtle but consequential ways.
The problem of withholding or failing to share raw data remains pervasive despite clear mandates from many scientific journals and funding agencies that promote open data as a standard for transparency and reproducibility. Studies have repeatedly documented a discrepancy between researchersβ stated intentions to share data and the reality of actual data availability. Researchers committing to data sharing during publication seldom follow through, leaving meta-analysts reliant on summary statistics, which inherently lack the granularity needed to fully validate or challenge reported outcomes. This shortfall reverberates through the entire evidence synthesis process.
In a policy forum article featured in PLOS Medicine, Saul Martin Rodriguez and colleagues at the Department of Laboratory Medicine, Karolinska Institutet, systematically explore the ramifications of this data transparency gap. They caution that the inability to access full datasets produces opaque analyses that complicate efforts to verify findings, reproduce studies, or understand underlying methodological nuances. These barriers undermine critical scientific valuesβreliability, transparency, and verificationβand potentially distort clinical and policy decisions that depend on synthesized evidence.
Historical precedents illustrate the danger of insufficient data access. The researchers specifically highlight hormone therapy for menopausal women, a high-profile instance where early meta-analyses, based on limited shared data, initially obscured risks associated with treatment. Only when detailed patient-level datasets eventually became available did the scientific community obtain a clearer, more accurate picture of the therapyβs adverse effects. This breakthrough recalibration led directly to amended medical guidelines, demonstrating how transparency can correct flawed understandings and safeguard public health.
Addressing the transparency crisis requires not only technical mandates but also cultural transformation within the research community. Enforcement of data sharing policies must be designed with consideration for ethical and legal constraints, such as patient privacy and intellectual property concerns. Moreover, the authors insist that transparency initiatives go beyond administrative checklists to fostering an environment where sharing data is recognized as a fundamental responsibility integral to research integrity.
Martin Rodriguez emphasizes that beyond the mechanics, transparency is a manifestation of scientific accountability. The trustworthiness of researchβand by extension public trust in scienceβdepends on open access to the underlying evidence. When data remain inaccessible, confidence erodes, and the risk of misguided healthcare interventions increases. Increasing openness mitigates these risks by enabling detailed scrutiny, replication efforts, and secondary analyses that can catch errors or biases earlier and more effectively.
The research article also identifies systemic obstacles impeding data availability, ranging from inadequate incentives for researchers to share, logistical burdens of data preparation, to concerns over misuse or misinterpretation of shared data. Addressing these challenges will require coordinated action from funding bodies, publishers, institutions, and researchers themselves to create infrastructure, reward systems, and legal frameworks that facilitate ethical and practical data sharing.
Collaboration with expert stakeholders, such as Professor David Moher of the Centre for Journalology, has strengthened the articleβs recommendations by drawing on established expertise in publication science and open research practices. Their joint insights offer concrete pathways for reforming meta-analytic methodology and promote strategies to leverage systematic reviews as truly transparent syntheses that can reliably guide healthcare policies.
This deep dive into data transparency exposes a critical vulnerability in the foundation of evidence-based medicine. The widespread lack of accessible underlying data in meta-analyses risks perpetuating incomplete or misleading conclusions with real-world consequences on patient care and resource allocation. Rectifying this crisis is imperative to uphold the scientific rigor and societal trust essential for advancing human health.
The articleβs publication comes at a time when open science ideals are gaining momentum worldwide. It presents a timely call to action that the scientific community must embrace if systematic reviews and meta-analyses are to fulfill their promise as gold standards of research evidence. By enforcing and normalizing comprehensive data sharing, researchers can drive greater accuracy, reproducibility, and ultimately, better health outcomes for populations globally.
Subject of Research: Data transparency issues in systematic reviews and meta-analyses and their impact on healthcare decision-making
Article Title: The data transparency crisis in research: Lessons from systematic reviews and meta-analyses
News Publication Date: 16-Jun-2026
Web References: https://doi.org/10.1371/journal.pmed.1005145
Image Credits: Photo by Karolinska Institutet
In 1992, mathematicians famously proved that seven βriffle shufflesβ β the kind where a player splits a deck of cards into two piles, then uses their thumbs to interleave them back together in a zipperlike motion β are enough to mix up the deck. When Dave Bayer and Persi Diaconis came up with this proof, they also revealed something surprising about what happens along the way: At firstβ¦
Hungarian mathematician Paul ErdΕsβ βplanar unit distance conjectureβ has stood since 1946 β but an AI disproof has made mathematicians pay attention.In the rapidly evolving field of computational pathology, artificial intelligence (AI) has demonstrated remarkable potential across a broad spectrum of diagnostic applications. Despite the impressive advances achieved by foundation and multimodal AI models, the translation of these technologies from experimental prototypes to everyday clinical practice remains sporadic and limited. This phenomenon, often described as the βadoption paradoxβ of computational pathology, highlights a profound disconnect between technical capability and practical clinical integration. New research published in LabMed Discovery offers a thorough examination of this paradox, proposing a robust framework to navigate the complexities of AI deployment in pathology.
At the core of the study lies a comprehensive three-stage maturity model that systematically categorizes the translational journey of AI in pathology. This framework details the progression from initial algorithmic development to full institutional adoption, articulating the critical barriers and enabling pathways that characterize each phase. Stage one, termed βAlgorithmic Capability,β centers on the development and validation of AI algorithms under retrospective conditions. However, here the barriers are significant: data heterogeneity, infrastructure instability, and the fragility introduced by variations in scanner technologies and data formats hinder robustness and reproducibility.
The study identifies that overcoming Stage oneβs obstacles requires an βinfrastructure-firstβ approach. This pathway prioritizes the establishment of vendor-neutral data formats and the implementation of automated quality control measures. By fostering multicenter benchmark baselines and utilizing federated learning techniques, developers can mitigate issues like scanner shift and manual quality-control dependencies, which currently limit algorithmic generalizability across diverse clinical environments.
Transitioning into Stage two, βSystem Integration,β the research highlights the necessity for prospective multisite validation as a gatekeeper to advance AI from the lab into live clinical workflows. Workflow disconnectsβsuch as cognitive rhythm mismatch, automation bias, and scenario-dependent latencyβpose significant barriers to embedding AI seamlessly in day-to-day pathology operations. These misalignments can erode clinical trust and reduce effective adoption, emphasizing the need for human-centered design.
To counter these challenges in Stage two, the report advocates for βworkflow-embedded intelligence,β a paradigm shift that integrates AI tools directly into the clinical workflow with features like triage prioritization, assistive decision layers, uncertainty visualization, and frictionless human override mechanisms. Such intelligent design ensures that AI functions as a collaborative partner rather than a disruptive force, supporting pathologistsβ expertise while providing transparent and actionable insights.
Stage three, βInstitutional Adoption,β focuses on achieving sustained use backed by demonstrable workflow benefits, viable reimbursement models, and comprehensive governance frameworks. This stage confronts deeply rooted barriers around institutional trust, interpretability, legal liability, and emerging risks associated with generative AI technologies. The complexity of securing organizational buy-in reflects the multifaceted nature of health systems, where regulatory accountability and risk management weigh heavily on technology acceptance.
To facilitate institutional trust and long-term integration, the authors underscore the importance of βadaptive governance.β This approach combines machine learning operations (MLOps) for continuous monitoring and maintenance, shadow deployments to validate real-world performance, and the collection of real-world evidence. Institutional policies must evolve to include robust validation procedures, liability frameworks, and protocols that address the nuances of generative AI risks, all crucial for sustainable clinical AI applications.
The maturity model also contextualizes current AI products within this framework, mapping approved and research-stage systems to their respective levels of advancement as of early 2026. This mapping provides a practical diagnostic tool, enabling stakeholders to identify specific bottlenecks and tailor interventions that promote progression toward routine clinical adoption. The ability to systematically understand why certain AI applications stall at particular stages offers a new lens for developers, policymakers, and clinical users alike.
Importantly, this comprehensive review sheds light on the broader evolution of pathology AI itself β tracing its trajectory from specialized task-specific deep learning models to the burgeoning domain of agentic systems capable of multimodal integration. Such advancement brings promise for more holistic and context-aware diagnostic solutions but simultaneously raises challenges in validation and governance that must be addressed proactively.
The authorsβ insights extend beyond technological innovation; they provide a strategic blueprint for reconciling AIβs groundbreaking potential with real-world healthcare complexities. By emphasizing infrastructure readiness, user-centric workflow design, and dynamic governance, this model aligns technical ambition with clinical realities, harmonizing diverse stakeholdersβ needs and expectations.
In conclusion, this study represents a pivotal contribution to computational pathologyβs ongoing AI revolution. Its detailed three-stage maturity framework captures the nuanced interplay of algorithm development, system embedding, and institutional acceptance, delineating actionable pathways to bridge the daunting gap between algorithmic promise and everyday clinical utility. As AI continues to reshape medical diagnostics, systematic and adaptive approaches like this will be essential to unlocking its full, sustainable impact on patient care worldwide.
Subject of Research: Artificial intelligence adoption in computational pathology and its clinical integration challenges
Article Title: Adoption paradox of artificial intelligence in computational pathology: a three-stage maturity model from algorithms to clinical integration
News Publication Date: 2-Jun-2026
Web References: http://dx.doi.org/10.1016/j.lmd.2026.100130
Image Credits: Lu Cai, Biwen Meng, Jie Huang, Guanyu Ding, Min Ju, Wenwen Wang, Shijie Deng, Liqin Lai, Jin Wang, Chunxue Yang, Miao Ruan, Shugong Xu, Chaofu Wang, Jingxin Liu, Qian Da
Keywords: Algorithms, Computational Pathology, Artificial Intelligence, Clinical Integration, Workflow Design, Infrastructure, Institutional Trust
For decades, physicists have pursued a goal that sounds nearly impossible: to build a clock that keeps time using an atom's nucleus rather than the electrons orbiting it.
Now, researchers have demonstrated the first functioning nuclear clock βͺββ¬ an advancement that could eventually lead to more robust timekeeping devices and new ways to search for dark matter and physics beyond the Standard Model.
"Having worked in this field for more than 15 years, it is just beautiful, how a very 'wild' idea such as manipulating an atomic nucleus with a laser has turned into reality," Thorsten Schumm, a professor of quantum metrology at the Vienna University of Technology and a member of the research team, told Live Science via email.
Today's most accurate clocks are optical atomic clocks, which measure the frequency of electrons jumping between different energy levels inside atoms. These clocks are so precise that they would lose less than a second over a 100 million years.
A nuclear clock works similarly, but it uses a transition within the nucleus itself, where the nucleus jumps between energy levels. Because the nucleus sits deep inside the atom, it's far less affected by external disturbances from things like electric or magnetic fields. According to Schumm, the nuclear transition can be 1,000 to 10,000 times less sensitive to environmental noise than atomic transitions are.
"This means that it would be easier to stabilize a nuclear clock over long periods of time," Jacob Higgins, a postdoctoral researcher at Northwestern University who previously worked on thorium clock experiments at JILA in Colorado but was not affiliated with the study, told Live Science in an email. "The transition used for the nuclear clock experiment has a higher quality factor than optical atomic clock transitions, which means that in principle, it can be measured more precisely given the same amount of measurement time."
Together, those advantages could allow nuclear clocks to outperform even today's best atomic clocks according to Higgins.
The nuclear clock relies on a rare isotope called thorium-229, whose nucleus contains an unusually low-energy excited state that can be manipulated with ultraviolet laser light.
For decades, scientists had suspected thorium had a low transition, but identifying and controlling it proved extremely challenging. Researchers spent years testing different thorium-containing materials, laser systems and detection methods before finally pinning down the transition.
"It was a long road," Higgins said.
One key advancement was the development of continuous-wave lasers that operate at the precise wavelength needed to excite the thorium nucleus. Before those lasers existed, researchers had to excite the nucleus and then wait several minutes for it to decay and emit a detectable signal. That process was too slow to build a practical clock.
"With the continuous lasers, we can measure the nucleus in absorption and get an immediate response, whether the laser is still at the right frequency (and if not, correct it back)," Schumm said. "Once we had that, it was 'just' implementing some electronics and atomization to have the clock stabilize itself to the nucleus."
Thanks to this set-up, the researchers kept the nuclear clock running continuously for 24 hours.
Unlike many optical atomic clocks, which require ultracold atoms to be suspended in a vacuum chamber, the thorium nuclei are embedded inside a crystal at room temperature.
Because the thorium transition remains stable inside a solid material, researchers may eventually be able to build compact clocks that are useful for navigation systems, telecommunication networks and data synchronization.
Some physicists are more excited about what nuclear clocks could reveal about fundamental physics, rather than the clocks' timekeeping abilities.
Atomic clocks primarily probe electromagnetic interactions involving electrons. Nuclear clocks, by contrast, are sensitive to the strong nuclear force, weak nuclear force and electromagnetism β three of the four fundamental forces of the universe, along with gravity. This can make them useful detectors of new physics, in a way.
"The nuclear clock is foremost a different clock, ticking on different fundamental physics mechanisms," Schumm said. "Essentially all modern theories beyond the standard model predict additional particles or 5th forces β¦ which can be probed with the nuclear clock in some parameter regime."
Thorium-229 is particularly intriguing because the energy difference between its two nuclear states results from a delicate balance between electromagnetic and nuclear forces. Because those large contributions nearly cancel each other out, even tiny changes in the underlying forces could shift the clock's frequency.
"So small shifts in these forces β like if the nucleus were to couple to certain forms of dark matter or if there were an oscillation of a fundamental constant β will be amplified in our measurement," Higgins said.
Researchers, including Higgins, have already used early versions of the clock to place constraints on some dark matter models, and they expect its sensitivity to improve as the technology itself gets better.
Although the first functioning nuclear clock is a major achievement, these timekeepers remain in their infancy. Scientists still need to gain a better understanding of how the thorium transition responds to factors like temperature and magnetic fields while developing more powerful and stable laser systems.
"I think it will be many years before the thorium clock can compete with today's best optical atomic clocks," Higgins said, "but we will learn a lot of new science on the pathway to getting there."
The restaurant problem reveals why human shortcuts can work surprisingly well.
Login