{
  "generated_at": "2026-09-16T21:12:27+00:00",
  "guides": [
    {
      "title": "How to read a new study without overclaiming",
      "url": "https://science.briefpulse.com/guides/read-a-new-study"
    }
  ],
  "publication": "science",
  "stories": [
    {
      "format": "brief",
      "headline": "Light-Controlled Microswimmers Switch Between Bacterial and Algal Propulsion",
      "published_at": "2026-09-16T21:07:16+00:00",
      "standfirst": "Physicists report a method to change a single artificial microswimmer's swimming style in real time, switching it between bacteria- and algae-inspired modes using light. The available summary does not quantify the sample, compare it with prior methods, or list limitations.",
      "url": "https://science.briefpulse.com/stories/science-light-controlled-microswimmers-switch-between-bacterial"
    },
    {
      "format": "brief",
      "headline": "Snow, monsoons and oceans nudge Earth's centre of mass each year, satellite tracking shows",
      "published_at": "2026-09-16T21:07:05+00:00",
      "standfirst": "Seasonal water movement shifts Earth's centre of mass back and forth by fractions of an inch relative to its geometric centre, according to NASA tracking. The source gives the direction and rough size of the wobble but no exact figure, no method detail and no stated uncertainty.",
      "url": "https://science.briefpulse.com/stories/science-snow-monsoons-oceans-nudge-earth"
    },
    {
      "format": "brief",
      "headline": "Fungi carried by bark beetles can neutralize spruce defense compounds, Max Planck team reports",
      "published_at": "2026-09-16T21:06:57+00:00",
      "standfirst": "A Max Planck Institute for Chemical Ecology team says certain fungi — delivered into trees by bark beetles — neutralize spruce defense compounds, which the source says makes beetle invasion easier. The announcement supplies no sample sizes, controls, fungal species names or published study detail, so the finding's reach is untested in what we have.",
      "url": "https://science.briefpulse.com/stories/science-fungi-carried-bark-beetles-can"
    },
    {
      "format": "brief",
      "headline": "Neutrino Laser Impossible Due to Physical Limits, MIT News Reports",
      "published_at": "2026-09-16T20:51:03+00:00",
      "standfirst": "MIT News reports that an earlier proposal for a neutrino laser cannot be achieved because of physical and fundamental limitations, according to scientists. The announcement provides no details on the analysis, so the certainty of the claim remains unclear.",
      "url": "https://science.briefpulse.com/stories/science-neutrino-laser-impossible-due-physical"
    },
    {
      "format": "brief",
      "headline": "Chariklo's two rings changed opacity within a few years, JWST observations show",
      "published_at": "2026-09-16T19:39:01+00:00",
      "standfirst": "The James Webb Space Telescope saw the inner ring of Chariklo, a small object orbiting between Saturn and Uranus, become more opaque while its outer ring became less opaque over just a few years. The report names no cause and gives no measurement figures, so the size and certainty of the change cannot yet be checked.",
      "url": "https://science.briefpulse.com/stories/science-chariklo-two-rings-changed-opacity"
    },
    {
      "format": "brief",
      "headline": "Simulations suggest a Venusian moon would have crashed into the planet",
      "published_at": "2026-09-16T19:13:59+00:00",
      "standfirst": "New computer simulations suggest Venus may once have had a moon that spiralled inward and hit the planet instead of drifting away like Earth's. The claim rests on modelling alone, and the material available here does not identify the researchers, the journal or the peer-review status of the work.",
      "url": "https://science.briefpulse.com/stories/science-simulations-suggest-venusian-moon-would"
    },
    {
      "format": "brief",
      "headline": "A model explains how track cyclists seem to start from rest with speed already on the clock",
      "published_at": "2026-09-16T19:03:09+00:00",
      "standfirst": "Some track cyclists look as though they begin a race already moving, in apparent violation of physics. A modelling study in Royal Society Open Science offers an explanation of the illusion, rather than a demonstration that any rule was broken.",
      "url": "https://science.briefpulse.com/stories/science-model-explains-track-cyclists-seem"
    },
    {
      "format": "brief",
      "headline": "NASA reports seasonal water shifts move Earth’s center of mass by fractions of an inch",
      "published_at": "2026-09-16T19:03:02+00:00",
      "standfirst": "NASA reports that seasonal water redistribution shifts Earth’s center of mass back and forth by fractions of an inch relative to its geometric center. The source does not detail the measurement method, time span, or uncertainty.",
      "url": "https://science.briefpulse.com/stories/science-nasa-reports-seasonal-water-shifts"
    },
    {
      "format": "brief",
      "headline": "NASA Reports Youngest Known Planet, Elias 2-24 b, Still in Its Natal Disk",
      "published_at": "2026-09-16T19:02:43+00:00",
      "standfirst": "NASA says astronomers have confirmed a planet less than a million years old, Elias 2-24 b, as the youngest known world. The announcement provides the record and the planet's setting, but the evidence supplied here does not detail how the age was measured or how it compares with other young candidates.",
      "url": "https://science.briefpulse.com/stories/science-nasa-reports-youngest-known-planet"
    },
    {
      "format": "brief",
      "headline": "Iraqi Arabic benchmark finds MSA tests saturate while dialect track separates models",
      "published_at": "2026-09-16T09:14:48+00:00",
      "standfirst": "A benchmark called Mizan evaluates large language models on Iraqi Arabic and the Iraqi civic context. In a pilot of 27 systems, the Modern Standard Arabic track saturated while the Iraqi track produced a consistent 14-18-point gap between models.",
      "url": "https://science.briefpulse.com/stories/science-iraqi-arabic-benchmark-finds-msa"
    },
    {
      "format": "brief",
      "headline": "National study finds more than half of U.S. adults experienced physical violence, nearly half sexual violence",
      "published_at": "2026-09-16T07:11:40+00:00",
      "standfirst": "A new national study reports that more than half of U.S. adults have experienced physical violence and nearly half have experienced sexual violence in their lifetimes. The study also links violence to economic hardship, mental health and substance use, but the available summary does not include methodological details.",
      "url": "https://science.briefpulse.com/stories/science-national-study-finds-more-than"
    },
    {
      "format": "brief",
      "headline": "High turnover among Tanzania's agro-dealers linked to farmer fertilizer adoption, study finds",
      "published_at": "2026-09-16T05:30:21+00:00",
      "standfirst": "A new study in the American Journal of Agricultural Economics examined Tanzania's agro-dealer sector and found that frequent turnover affects farmer perceptions of product quality, with implications for market functioning.",
      "url": "https://science.briefpulse.com/stories/science-high-turnover-among-tanzania-agro-dealers"
    },
    {
      "format": "brief",
      "headline": "Small correction module fixes 53% of frozen model errors in limited study",
      "published_at": "2026-09-16T05:06:06+00:00",
      "standfirst": "A new study tests whether a small correction module can fix errors in a frozen language model without degrading its base abilities. It reports 53.3% error correction on a 60-question exam with no measured degradation on small tested benchmarks, but the evidence is from a preprint and limited evaluations.",
      "url": "https://science.briefpulse.com/stories/science-small-correction-module-fixes-frozen"
    },
    {
      "format": "brief",
      "headline": "AI vs. traditional methods: a new meta-analysis finds a mixed record",
      "published_at": "2026-09-16T04:23:09+00:00",
      "standfirst": "A preprint analyzing 2,507 head-to-head comparisons between AI and other scientific methods across 27 disciplines finds a mixed record: AI often beats traditional statistics but at higher cost, and while it has improved against scientific computing since 2020, it still underperforms in many cases.",
      "url": "https://science.briefpulse.com/stories/science-traditional-methods-meta-analysis-finds-mixed"
    },
    {
      "format": "brief",
      "headline": "CADWorld benchmark tests AI agents on long-horizon FreeCAD design tasks",
      "published_at": "2026-09-16T04:22:43+00:00",
      "standfirst": "A benchmark called CADWorld evaluates computer-use agents on long-horizon mechanical CAD tasks in FreeCAD, with success measured by executable checks on saved artifacts. The strongest of seven agents solved 17.5% of tasks, against an 87.0% expert reference pass.",
      "url": "https://science.briefpulse.com/stories/science-cadworld-benchmark-tests-agents-long-horizon"
    },
    {
      "format": "brief",
      "headline": "Narrative Review Maps Eight Governance Risks in LLM-Enabled Geospatial AI",
      "published_at": "2026-09-16T04:22:14+00:00",
      "standfirst": "A narrative review posted to arXiv synthesizes ethical and privacy risks in geospatial AI powered by large language models. It identifies eight recurring issues and proposes a governance-aware architecture, but notes that field-tested evaluations of such controls remain limited.",
      "url": "https://science.briefpulse.com/stories/science-narrative-review-maps-eight-governance"
    },
    {
      "format": "brief",
      "headline": "Calibrated router for disaggregated LLM serving shows goodput gains in measured tests",
      "published_at": "2026-09-16T04:12:37+00:00",
      "standfirst": "A preprint reports a router that estimates completion time per instance using prompt length, predicted output length, KV cache pressure, and SLO class. Tested on eight NVIDIA A40 GPUs, it achieved the highest mean goodput across three bursty traces.",
      "url": "https://science.briefpulse.com/stories/science-calibrated-router-disaggregated-llm-serving"
    },
    {
      "format": "brief",
      "headline": "Preprint proposes AI routing for shooting evacuations, tested only in simulation",
      "published_at": "2026-09-16T04:11:56+00:00",
      "standfirst": "A new arXiv preprint describes GPEvac, a graph-neural-network system that computes evacuation routes during shooting events. The work reports simulation results, not real-world deployments.",
      "url": "https://science.briefpulse.com/stories/science-preprint-proposes-routing-shooting-evacuations"
    },
    {
      "format": "brief",
      "headline": "AI's biosecurity risk depends on users and lab access, review says",
      "published_at": "2026-09-16T04:07:31+00:00",
      "standfirst": "A review of artificial intelligence and biosecurity argues that AI can help with digital biological tasks more than physical ones, and that risk depends on who is using it and what safeguards exist.",
      "url": "https://science.briefpulse.com/stories/science-biosecurity-risk-depends-users-lab"
    },
    {
      "format": "brief",
      "headline": "Same Rain, Different Rivers: Catchments Vary in Turning Rainfall into Flow",
      "published_at": "2026-09-15T23:31:37+00:00",
      "standfirst": "A new study in Nature Water finds that the same amount of rain on the same landscape can send very different amounts of water into rivers at different times. The research provides a global picture of how consistently catchments turn rainfall into river flow.",
      "url": "https://science.briefpulse.com/stories/science-same-rain-different-rivers-catchments"
    },
    {
      "format": "brief",
      "headline": "Webb finds brown dwarfs about twice Jupiter's mass in IC 348",
      "published_at": "2026-09-15T23:22:03+00:00",
      "standfirst": "Astronomers using the James Webb Space Telescope searched the nearby star-forming region IC 348 and found brown dwarfs about twice Jupiter's mass, according to the report. The report gives the mass but little detail beyond it.",
      "url": "https://science.briefpulse.com/stories/science-webb-finds-brown-dwarfs-twice"
    },
    {
      "format": "brief",
      "headline": "Ancient wall discovered at Oued Beht in Morocco",
      "published_at": "2026-09-15T23:21:34+00:00",
      "standfirst": "Archaeologists have found a substantial built structure, including an ancient wall, at Oued Beht in northwestern Morocco. The discovery suggests second-millennium BC communities there were shaping their environment and linked to Mediterranean networks.",
      "url": "https://science.briefpulse.com/stories/science-ancient-wall-discovered-oued-beht"
    },
    {
      "format": "brief",
      "headline": "Canadians Reject Vandalism and Violence as Protest, Study Finds, but Tolerance Varies",
      "published_at": "2026-09-15T23:21:17+00:00",
      "standfirst": "A new study reports that Canadians overall do not support vandalism or violence as forms of environmental protest, while tolerance for disruptive tactics depends on several factors.",
      "url": "https://science.briefpulse.com/stories/science-canadians-reject-vandalism-violence-protest"
    },
    {
      "format": "brief",
      "headline": "Irrigation's greenhouse gas benefits far outweigh emissions costs, study finds",
      "published_at": "2026-09-15T20:20:21+00:00",
      "standfirst": "A new study from Colorado State University finds that irrigating U.S. crops prevents higher greenhouse gas emissions by avoiding land conversion. According to a brief report, if irrigation were suddenly shut off, the land-use change required to make up for lost yields would produce 363 times more emissions than irrigation emits.",
      "url": "https://science.briefpulse.com/stories/science-irrigation-greenhouse-gas-benefits-far"
    },
    {
      "format": "brief",
      "headline": "Roman road network analyzed in most detailed quantitative picture yet",
      "published_at": "2026-09-15T20:19:56+00:00",
      "standfirst": "A multidisciplinary team analyzed a dataset of almost 300,000 kilometers of Roman roads across the former empire, producing what the source calls the most detailed quantitative picture yet of the network's structure. The evidence provided is a short summary and does not detail the specific findings or how certainty was assessed.",
      "url": "https://science.briefpulse.com/stories/science-roman-road-network-analyzed-most"
    },
    {
      "format": "brief",
      "headline": "Pilot study asks experts to forecast AI R&D acceleration impacts",
      "published_at": "2026-09-15T19:17:48+00:00",
      "standfirst": "METR and the Forecasting Research Institute ran a pilot survey of 8 AI experts and 10 superforecasters. The survey asked them to estimate the likelihood of rapid acceleration and transformative impacts conditional on hypothetical evaluation results.",
      "url": "https://science.briefpulse.com/stories/science-pilot-study-asks-experts-forecast"
    },
    {
      "format": "brief",
      "headline": "AI long-task benchmark post warns some claims are out of date",
      "published_at": "2026-09-15T18:58:24+00:00",
      "standfirst": "A research blog post titled 'Measuring AI Ability to Complete Long Software Tasks' carries a banner saying some of its text and figures are out of date.",
      "url": "https://science.briefpulse.com/stories/science-long-task-benchmark-post-warns-some"
    },
    {
      "format": "brief",
      "headline": "Paper models when AI feedback could lead to self-sustaining acceleration",
      "published_at": "2026-09-15T17:41:03+00:00",
      "standfirst": "A new paper coauthored by Parker, Tom, and seven other economists presents simple models of how AI may accelerate AI R&D. It focuses on the strength of feedback effects and whether they could cause self-sustaining acceleration.",
      "url": "https://science.briefpulse.com/stories/science-paper-models-feedback-could-lead"
    },
    {
      "format": "brief",
      "headline": "Black hole winds 100 times more powerful than thought, XRISM observations suggest",
      "published_at": "2026-09-15T15:07:26+00:00",
      "standfirst": "Astronomers have found that winds from a supermassive black hole are about 100 times more powerful than previously thought, with turbulence spreading roughly 300,000 light-years. The report does not detail the study's methods or how certain the result is.",
      "url": "https://science.briefpulse.com/stories/science-black-hole-winds-100-times"
    },
    {
      "format": "brief",
      "headline": "Timing changes to farmland conservation may help UK wading birds, research says",
      "published_at": "2026-09-15T14:53:01+00:00",
      "standfirst": "New research from the University of Aberdeen and Edinburgh Napier University suggests that simple changes to the timing of conservation measures on farmland could significantly improve protection for some of the U.K.'s most rapidly declining wading bird species. The available summary does not specify which species, what timing changes, or the size of the effect.",
      "url": "https://science.briefpulse.com/stories/science-timing-changes-farmland-conservation-may"
    },
    {
      "format": "brief",
      "headline": "Webb Telescope Finds Brown Dwarfs at Twice Jupiter's Mass in IC 348",
      "published_at": "2026-09-15T14:52:48+00:00",
      "standfirst": "NASA's James Webb Space Telescope has captured a large image of the star-forming region IC 348. Astronomers used it to search for brown dwarfs and discovered some with masses only twice that of Jupiter.",
      "url": "https://science.briefpulse.com/stories/science-webb-telescope-finds-brown-dwarfs"
    },
    {
      "format": "brief",
      "headline": "Little fire ants in lab tests chose cooked beef and sunflower seeds over tuna",
      "published_at": "2026-09-15T14:52:34+00:00",
      "standfirst": "In laboratory food-choice tests, little fire ants overwhelmingly picked cooked beef and sunflower seeds over water-packed tuna, researchers at the University of Hawaiʻi at Mānoa report. The result points to a possible change in how the pests are detected and baited.",
      "url": "https://science.briefpulse.com/stories/science-little-fire-ants-lab-tests"
    },
    {
      "format": "brief",
      "headline": "New research estimates much larger pre-European Indigenous Australian population and 93% loss after 1788",
      "published_at": "2026-09-15T11:49:47+00:00",
      "standfirst": "New research indicates that Australia's Indigenous population before European arrival was much larger than previously believed, and that an estimated 93% was lost within 80 years of the First Fleet's 1788 arrival. The source does not provide the absolute population numbers or methods behind the estimate.",
      "url": "https://science.briefpulse.com/stories/science-research-estimates-much-larger-pre-european"
    },
    {
      "format": "brief",
      "headline": "NGC 1068's Black Hole Powers a Hidden, Violent Gas Outflow",
      "published_at": "2026-09-15T11:49:35+00:00",
      "standfirst": "Astronomers report that the nearby galaxy NGC 1068 has a black hole driving a more powerful and previously hidden gas outflow than realized. The finding comes from combining new infrared observations with existing data.",
      "url": "https://science.briefpulse.com/stories/science-ngc-1068-black-hole-powers"
    },
    {
      "format": "brief",
      "headline": "Web agent OdoBot cuts token use by up to 80% in Canvas LMS tests, abstract says",
      "published_at": "2026-09-15T06:06:08+00:00",
      "standfirst": "A new web-agent architecture, OdoBot, is described in an arXiv abstract as completing tasks with fewer tokens than two competitor agents. The abstract reports experiments on 45 Canvas LMS tasks but does not provide peer-review or full methodological details.",
      "url": "https://science.briefpulse.com/stories/science-web-agent-odobot-cuts-token"
    },
    {
      "format": "brief",
      "headline": "GAVEL compares clinical timelines against case reports; merged timelines preferred in 77.0% of comparisons",
      "published_at": "2026-09-15T05:56:29+00:00",
      "standfirst": "A preprint describes GAVEL, an LLM judge protocol that compares two clinical timelines with the case report and returns a discrepancy type, verdict, and report passage for each difference. It reports that merged timelines were preferred in 77.0% of comparisons across 126 reports.",
      "url": "https://science.briefpulse.com/stories/science-gavel-compares-clinical-timelines-against"
    },
    {
      "format": "brief",
      "headline": "Two-stage method splits convergence and diversity in multi-objective optimisation",
      "published_at": "2026-09-15T05:56:10+00:00",
      "standfirst": "A new arXiv preprint proposes a 'converge-then-diversify' approach for multi-objective Bayesian optimisation, but the available evidence is only the paper's abstract.",
      "url": "https://science.briefpulse.com/stories/science-two-stage-method-splits-convergence-diversity"
    },
    {
      "format": "brief",
      "headline": "Researchers Introduce OrchSLM to Probe Small Language Model Orchestration",
      "published_at": "2026-09-15T05:55:51+00:00",
      "standfirst": "A new preprint introduces OrchSLM, a routing framework for coordinating small language models without direct interaction. The abstract describes the framework but does not report specific performance results, so the finding's practical impact remains unclear.",
      "url": "https://science.briefpulse.com/stories/science-researchers-introduce-orchslm-probe-small"
    },
    {
      "format": "brief",
      "headline": "TimeThink preprint claims synthetic training improves timeseries reasoning in AI models",
      "published_at": "2026-09-15T05:45:20+00:00",
      "standfirst": "A new arXiv preprint describes TimeThink, a synthetic-data framework for training timeseries language models to reason about patterns like trend and seasonality. The abstract claims it outperforms baselines, but it provides no numerical results or peer-review details, so the finding is not yet independently verified.",
      "url": "https://science.briefpulse.com/stories/science-timethink-preprint-claims-synthetic-training"
    },
    {
      "format": "brief",
      "headline": "AI agent LabAgent claims top rank in four life-science tasks, per arXiv abstract",
      "published_at": "2026-09-15T05:44:53+00:00",
      "standfirst": "An arXiv abstract describes LabAgent, a system for reproducing and extending lab methods. The evidence available is limited to that abstract, so the results are not yet independently confirmed here.",
      "url": "https://science.briefpulse.com/stories/science-agent-labagent-claims-top-rank"
    },
    {
      "format": "brief",
      "headline": "Preprint: Zero-shot LLM agents match RL on agricultural tasks, adapt better to weather shifts",
      "published_at": "2026-09-15T05:35:56+00:00",
      "standfirst": "A preprint on arXiv reports that a multi-agent framework using large language models can manage long-term physical tasks in a zero-shot manner. The abstract claims the agents matched reinforcement learning agents under the same weather pattern and adapted better under a shifted environment, but the work has not been peer-reviewed and only the abstract is available.",
      "url": "https://science.briefpulse.com/stories/science-preprint-zero-shot-llm-agents-match"
    },
    {
      "format": "brief",
      "headline": "AI judges improve AI-written patent drafts, but match a human attorney only partly",
      "published_at": "2026-09-15T05:35:40+00:00",
      "standfirst": "A new arXiv preprint describes a patent-drafting testbed in which a separate AI judge scores drafts and feeds back revisions. The reported gains are measured largely by that same judge, and its agreement with a professional patent attorney was meaningful but depended heavily on which metric was used.",
      "url": "https://science.briefpulse.com/stories/science-judges-improve-ai-written-patent-drafts"
    },
    {
      "format": "brief",
      "headline": "Open 7B model ZGCM-1 claims competitive math and agentic search results",
      "published_at": "2026-09-15T05:35:05+00:00",
      "standfirst": "A newly posted abstract describes ZGCM-1, a 7B open model for math and agentic search. The claims come from the authors' abstract, not from independent replication.",
      "url": "https://science.briefpulse.com/stories/science-open-model-zgcm-1-claims-competitive"
    },
    {
      "format": "brief",
      "headline": "Study Finds Anger, Hope, Compassion Do Little to Spur Climate Donations",
      "published_at": "2026-09-14T23:30:15+00:00",
      "standfirst": "A new study that asked participants to donate money found that anger, hope, and compassion did little to increase climate donations. The available evidence offers only a brief summary of the research.",
      "url": "https://science.briefpulse.com/stories/science-study-finds-anger-hope-compassion"
    },
    {
      "format": "brief",
      "headline": "Sex has little impact on monk parakeet fights and dominance, study finds",
      "published_at": "2026-09-14T23:30:04+00:00",
      "standfirst": "Researchers at the University of Cincinnati and Duke University studied monk parakeets and found that a bird's sex had little to no effect on whether it started fights, was targeted, or rose in the dominance hierarchy.",
      "url": "https://science.briefpulse.com/stories/science-sex-has-little-impact-monk"
    },
    {
      "format": "brief",
      "headline": "NYC-area methane traced mainly to natural gas use, study finds",
      "published_at": "2026-09-14T20:39:00+00:00",
      "standfirst": "A new study in Atmospheric Chemistry and Physics reports that methane emissions across the New York City metropolitan area are largely driven by natural gas use. The authors suggest cutting that waste could also save consumers money.",
      "url": "https://science.briefpulse.com/stories/science-nyc-area-methane-traced-mainly-natural"
    },
    {
      "format": "brief",
      "headline": "Pigment analysis of painting signed 'L daVinci' finds Renaissance materials and later alterations",
      "published_at": "2026-09-14T20:29:20+00:00",
      "standfirst": "Researchers at Uppsala University analyzed pigments in a painting signed 'L daVinci.' Most are consistent with Renaissance materials, but one more recent pigment appears linked to later alterations.",
      "url": "https://science.briefpulse.com/stories/science-pigment-analysis-painting-signed-davinci"
    },
    {
      "format": "brief",
      "headline": "Peptide position inside liposomes steers gold nanoparticle shape",
      "published_at": "2026-09-14T20:28:07+00:00",
      "standfirst": "A research team from the Institute of Science Tokyo reports that where biomineralization peptides sit within liposomes changes whether gold nanoparticles grow branched or spherical. The work suggests a way to control nanoscale reaction environments.",
      "url": "https://science.briefpulse.com/stories/science-peptide-position-inside-liposomes-steers"
    },
    {
      "format": "brief",
      "headline": "Small study links 8-hour eating window to better blood sugar control in Type 1 diabetes",
      "published_at": "2026-09-14T17:43:30+00:00",
      "standfirst": "A small study of adults with Type 1 diabetes and obesity found that eating only between noon and 8 p.m. improved blood sugar control without increasing serious complications. The findings are the first evidence that time-restricted eating could become another tool for managing the condition.",
      "url": "https://science.briefpulse.com/stories/science-small-study-links-8-hour-eating"
    },
    {
      "format": "brief",
      "headline": "Michigan Scholarship Program Tied to 4% Increase in On-Time Degree Attainment",
      "published_at": "2026-09-14T17:33:12+00:00",
      "standfirst": "A study co-authored by researchers from the University of Michigan, Harvard University and elsewhere found that the HAIL Scholarship program increased on-time bachelor's degree attainment by 4% for low-income Michigan students. The evidence provided is limited to this single result.",
      "url": "https://science.briefpulse.com/stories/science-michigan-scholarship-program-tied-increase"
    },
    {
      "format": "brief",
      "headline": "Australian social isolation nearly doubled over two decades, University of Sydney research reports",
      "published_at": "2026-09-14T17:32:18+00:00",
      "standfirst": "A University of Sydney research summary reports that social isolation rates in Australia nearly doubled over 23 years, with loneliness rising more gradually. The news report offers limited methodological detail.",
      "url": "https://science.briefpulse.com/stories/science-australian-social-isolation-nearly-doubled"
    },
    {
      "format": "brief",
      "headline": "Snail Shell Rings May Hold Records of Prehistoric Cyclones, Study Finds",
      "published_at": "2026-09-14T14:32:02+00:00",
      "standfirst": "A University of Queensland study has examined growth rings on the shell of a Biggenden Banded Snail, suggesting that snail shells could serve as weather time capsules for mapping prehistoric cyclones.",
      "url": "https://science.briefpulse.com/stories/science-snail-shell-rings-may-hold"
    },
    {
      "format": "brief",
      "headline": "New method matches concept-learning benchmarks without hand-crafted loss, preprint says",
      "published_at": "2026-09-14T06:14:39+00:00",
      "standfirst": "A preprint on arXiv describes Soft-PNet, a neuro-symbolic method that removes a hand-crafted loss and instead uses a Metropolis walk over cached symbolic solutions. It reports matching existing methods on three benchmarks under scarce supervision.",
      "url": "https://science.briefpulse.com/stories/science-method-matches-concept-learning-benchmarks-without"
    },
    {
      "format": "brief",
      "headline": "Synthetic office-email benchmark leaves AI agents short of 80%",
      "published_at": "2026-09-14T06:01:31+00:00",
      "standfirst": "A preprint describes an automated pipeline that generates synthetic workplace-email datasets and reports that standard agentic baselines score below 80% on them. The evidence is a preprint built on simulated data, so the numbers describe one constructed test rather than real enterprise performance.",
      "url": "https://science.briefpulse.com/stories/science-synthetic-office-email-benchmark-leaves-agents"
    },
    {
      "format": "brief",
      "headline": "New benchmark and method aim to forecast meeting continuations",
      "published_at": "2026-09-14T06:01:16+00:00",
      "standfirst": "A preprint describes a benchmark built from 2,207 real meetings and a method called GLARE that generates plausible meeting continuations. The results are from human evaluations, but the work has not yet been peer-reviewed.",
      "url": "https://science.briefpulse.com/stories/science-benchmark-method-aim-forecast-meeting"
    },
    {
      "format": "brief",
      "headline": "Study probes when LLM draft-verify-revise pipelines disagree on 'previous'",
      "published_at": "2026-09-14T05:50:43+00:00",
      "standfirst": "A preprint abstract reports that in draft-verify-revise LLM pipelines, a context-dependent expression such as 'previous' can be resolved differently by different stages, producing a deictic shift. The study used a synthetic dataset of 10 base examples in three conditions and tested six models across 21 reasoning effort configurations.",
      "url": "https://science.briefpulse.com/stories/science-study-probes-llm-draft-verify-revise-pipelines"
    },
    {
      "format": "brief",
      "headline": "A smaller neural operator reports better wave-rollout accuracy than a larger baseline on its authors' benchmark",
      "published_at": "2026-09-14T05:49:06+00:00",
      "standfirst": "A preprint introduces DU-NO, a neural operator for phase-resolving wave modeling, and reports it beats a larger baseline on the authors' own benchmark. The evidence is a single unreviewed preprint, so treat the numbers as the authors' reported results, not settled ones.",
      "url": "https://science.briefpulse.com/stories/science-smaller-neural-operator-reports-better"
    },
    {
      "format": "brief",
      "headline": "Cardiac AI model trained to read several test types at once, preprint reports",
      "published_at": "2026-09-14T05:28:05+00:00",
      "standfirst": "A preprint describes a model that learns from ECG, echocardiography, chest radiographs and clinical variables together rather than one at a time. The claims come from the authors' own benchmarks and have not been through peer review, so treat the numbers as provisional.",
      "url": "https://science.briefpulse.com/stories/science-cardiac-model-trained-read-several"
    },
    {
      "format": "brief",
      "headline": "Vendor-Native Coding Harnesses Show No Clear Average Advantage in Paired Test",
      "published_at": "2026-09-14T05:21:13+00:00",
      "standfirst": "A new arXiv paper compares agentic coding harnesses paired with the same models on a private, contamination-controlled suite. It reports no resolved average advantage for either harness, with opposite results across task types and unresolved cost ordering.",
      "url": "https://science.briefpulse.com/stories/science-vendor-native-coding-harnesses-show-clear"
    },
    {
      "format": "brief",
      "headline": "A 35B 'co-work' model lands at the cheap end of a cost-performance curve, its authors report",
      "published_at": "2026-09-14T05:19:05+00:00",
      "standfirst": "A preprint describes Occamy-1.0, a 35-billion-parameter agent model built by further training an existing checkpoint, and reports it is competitive with larger systems on the authors' own benchmarks. The claims rest on the team's stated evaluation and pricing protocol, not independent testing.",
      "url": "https://science.briefpulse.com/stories/science-35b-co-work-model-lands-cheap"
    },
    {
      "format": "brief",
      "headline": "JWST Finds No Exomoon, Ruling Out Tiny Moons Around One Exoplanet",
      "published_at": "2026-09-14T01:29:32+00:00",
      "standfirst": "The James Webb Space Telescope has yet to confirm an exomoon. A new result rules out tiny moons around a single exoplanet, a non-detection the report calls great news.",
      "url": "https://science.briefpulse.com/stories/science-jwst-finds-exomoon-ruling-tiny"
    },
    {
      "format": "brief",
      "headline": "Companion Star, Not Supernova, May Explain Distant Star's Outbursts",
      "published_at": "2026-09-13T22:23:47+00:00",
      "standfirst": "A distant star that astronomers thought would soon explode as a supernova may instead be flaring because of a previously unseen companion, new research suggests.",
      "url": "https://science.briefpulse.com/stories/science-companion-star-not-supernova-may"
    },
    {
      "format": "brief",
      "headline": "Luminescence dating pushes Mammoth Site age back more than 100,000 years",
      "published_at": "2026-09-13T19:23:06+00:00",
      "standfirst": "New dates published in Quaternary Research suggest the mammoths found in a South Dakota sinkhole were more than 160,000 years old, not roughly 26,000 years old as long accepted.",
      "url": "https://science.briefpulse.com/stories/science-luminescence-dating-pushes-mammoth-site"
    },
    {
      "format": "brief",
      "headline": "MemBrain v2 automates 3D membrane mapping, cutting weeks of work to hours",
      "published_at": "2026-09-13T19:22:47+00:00",
      "standfirst": "A team from Helmholtz Munich, the Technical University of Munich and the Biozentrum of the University of Basel has built MemBrain v2, an AI tool that automates 3D membrane mapping. The study is published in Nature Methods.",
      "url": "https://science.briefpulse.com/stories/science-membrain-automates-membrane-mapping-cutting"
    },
    {
      "format": "brief",
      "headline": "Semaglutide linked to fewer asthma attacks and COPD flare-ups in real-world study",
      "published_at": "2026-09-13T00:24:33+00:00",
      "standfirst": "A large real-world study linked semaglutide, the drug in Ozempic and Wegovy, to nearly 40% fewer asthma attacks and about 20% fewer COPD flare-ups. Researchers say clinical trials are needed to confirm the findings.",
      "url": "https://science.briefpulse.com/stories/science-semaglutide-linked-fewer-asthma-attacks"
    },
    {
      "format": "brief",
      "headline": "Daily sugary drinks linked to higher stomach cancer risk in large observational study",
      "published_at": "2026-09-13T00:24:18+00:00",
      "standfirst": "A decades-long study of more than 112,000 U.S. adults found that those who drank at least one sugar-sweetened beverage a day had roughly 2.45 times the risk of gastric cancer compared with those who rarely consumed them. The researchers stress the findings do not prove sugary beverages cause cancer.",
      "url": "https://science.briefpulse.com/stories/science-daily-sugary-drinks-linked-higher"
    },
    {
      "format": "brief",
      "headline": "Columbia study: embryo base editing reaches all cells but brings unpredictable changes",
      "published_at": "2026-09-13T00:20:02+00:00",
      "standfirst": "A Columbia University study reports that new gene-editing techniques can accurately edit genes in human embryos, but the work also uncovered risks that currently rule out clinical use.",
      "url": "https://science.briefpulse.com/stories/science-columbia-study-embryo-base-editing"
    },
    {
      "format": "brief",
      "headline": "Simulations suggest Southern Ocean could flip from carbon sink to carbon source",
      "published_at": "2026-09-13T00:19:50+00:00",
      "standfirst": "A study published in Science Advances finds that climate simulations suggest the Southern Ocean around Antarctica may switch from an important carbon sink to a prolific source of carbon as CO2 levels are reduced.",
      "url": "https://science.briefpulse.com/stories/science-simulations-suggest-southern-ocean-could"
    },
    {
      "format": "brief",
      "headline": "Hayabusa2 conducts first laser-ranging experiment with asteroid Torifune",
      "published_at": "2026-09-13T00:19:35+00:00",
      "standfirst": "JAXA says its Hayabusa2 spacecraft carried out the first successful laser-ranging experiment with an asteroid during a July 5 flyby of Torifune, the first target of its extended mission.",
      "url": "https://science.briefpulse.com/stories/science-hayabusa2-conducts-first-laser-ranging-experiment"
    },
    {
      "format": "brief",
      "headline": "Preprint proposes intent-driven privacy filtering for LLM prompts",
      "published_at": "2026-09-12T21:38:03+00:00",
      "standfirst": "A new arXiv preprint proposes an intent-driven framework that strips sensitive details from prompts before they reach a language model, using a distilled model called Veilmind-4B. The authors say it preserves more response utility than existing privacy baselines.",
      "url": "https://science.briefpulse.com/stories/science-preprint-proposes-intent-driven-privacy-filtering"
    },
    {
      "format": "brief",
      "headline": "Preprint scheduler cuts agentic LLM workflow tail latency by up to 3.50×",
      "published_at": "2026-09-12T07:53:10+00:00",
      "standfirst": "A new arXiv paper proposes releasing model turns only when tail risk is low, instead of the moment they are ready. On real software-engineering traces it matched eager release under light load and cut P95 workflow flow time under contention by up to 3.50×.",
      "url": "https://science.briefpulse.com/stories/science-preprint-scheduler-cuts-agentic-llm"
    },
    {
      "format": "brief",
      "headline": "Preprint proposes audit protocol for AI update admission, tested only in simulation",
      "published_at": "2026-09-12T07:50:57+00:00",
      "standfirst": "A new preprint argues that update admission for continually learning AI agents should be judged by both error control and retained learning. In a simulated one-step pushing task with 32 seeds, a paired-binomial check admitted 31.6% of updates at 2,000 episodes per stage, while a range-based gate admitted none; physical-robot validation remains open.",
      "url": "https://science.briefpulse.com/stories/science-preprint-proposes-audit-protocol-update"
    },
    {
      "format": "brief",
      "headline": "Clinical calculator AI: code solver lifts 32B model, not 7B, preprint finds",
      "published_at": "2026-09-12T07:48:01+00:00",
      "standfirst": "A preprint tests whether language models do clinical arithmetic better by writing Python for a restricted executor to run instead of calculating directly. It reports a gain for a 32B model but not a 7B one, and says verified formulas remain essential.",
      "url": "https://science.briefpulse.com/stories/science-clinical-calculator-code-solver-lifts"
    },
    {
      "format": "brief",
      "headline": "Probabilistic Focal Search preprint cuts node expansions when FOCAL admission stalls",
      "published_at": "2026-09-12T07:44:04+00:00",
      "standfirst": "An arXiv preprint introduces Probabilistic Focal Search, a variant of Focal Search that sometimes expands a minimum-f node instead of the heuristic-guided choice. In benchmark tests, the largest gains appeared when long f_min plateaus delayed useful FOCAL admissions.",
      "url": "https://science.briefpulse.com/stories/science-probabilistic-focal-search-preprint-cuts"
    }
  ]
}