- These are the states that rank highest and lowest for women’s health in new report
- Long Covid risk has dropped over time but remains substantial, study shows
- Are cold showers good for you? What the science says
- Two killed as walkway collapses at Italy’s notorious ‘Gomorrah’ slum
- AI models fed AI-generated data quickly spew nonsense
- Exclusive: the Trump administration demoted this climate scientist — now she wants reform
These are the states that rank highest and lowest for women’s health in new report
這些是新報告中女性健康排名最高和最低的州
Health
In their inaugural state-by-state analysis on women’s health, researchers at the Commonwealth Fund, a private foundation focused on health issues, collected data on health care quality, outcomes and access for women in the US. The data came from several sources, including the US Centers for Disease Control and Prevention, and the researchers evaluated states on 32 specific metrics, scoring them relative to each other. The report, released Thursday, reveals that states in the Northeast scored highest. Massachusetts came out on top as “the best-performing health system for women overall,” with Vermont, Rhode Island, Connecticut and New Hampshire rounding out the top five. “While some states undoubtedly are championing women’s continued access to vital health and reproductive services, many others are failing to ensure that women can get and afford the health care they need. This failure is having a disproportionate impact on women of color and women with low incomes,” he said. “My hope is that policymakers can use these findings to identify and address gaps in care, guaranteeing that all women across the United States can live healthy lives with access to quality, affordable care – no matter where they live or what their background is.”
---from CNN
在針對女性健康的首次州際分析中，專注於健康問題的私人基金會--聯邦基金會的研究人員收集了有關美國女性醫療保健品質、結果和獲得機會的數據。這些數據來自多個來源，包括美國疾病管制與預防中心，研究人員根據 32 個具體指標對各州進行了評估，並對它們進行相對評分。 週四發布的報告顯示，東北部各州得分最高。麻薩諸塞州位居「女性整體表現最佳的衛生系統」榜首，佛蒙特州、羅德島州、康乃狄克州和新罕布夏州也躋身前五名。 「雖然一些州無疑支持婦女繼續獲得重要的健康和生殖服務，但許多其他州未能確保婦女能夠獲得並負擔得起她們所需的醫療保健。這種失當對有色人種女性和低收入女性產生了不成比例的影響，」他說。「我希望政策制定者能夠利用這些發現來確定和解決護理方面的差距，保證美國所有女性都能過上健康的生活，獲得優質、負擔得起的照護——無論她們住在哪里或背景是什麼。
---摘錄翻譯自CNN
Long Covid risk has dropped over time but remains substantial, study shows
研究顯示，長期感染新冠病毒的風險隨著時間的推移有所下降，但仍很大
Health
With a summer wave of Covid-19 infections sweeping the country, a timely new study has looked at the risk of getting long Covid and whether those odds have changed over time. It found that the likelihood of developing long Covid has dropped since the start of the pandemic but remains substantial, especially for people who aren’t vaccinated against the coronavirus. About 7% of American adults, roughly 18 million people, have ever had long Covid, according to an analysis by the federal Agency for Healthcare Research and Quality that was published in June. Harvard economist David Cutler estimated in 2022 that the total cost of long Covid to the nation was $3.7 trillion, or 17% of the country’s pre-Covid gross domestic product. The new study, which was published Wednesday in the New England Journal of Medicine, suggests that the human and financial toll will only grow. The investigation leaned on computers and advanced machine learning to sift through the data in millions of medical records maintained by the Department of Veterans Affairs.
---from CNN
隨著夏季 Covid-19 感染浪潮席捲全國，一項及時的新研究探討了新冠長期症狀的風險以及這些機率是否隨著時間的推移而變化。 研究發現。自大流行開始以來，新冠長期症狀的可能性有所下降，但仍然很大，特別是對於沒有接種冠狀病毒疫苗的人來說。 根據聯邦衛生保健研究和質量監督局6 月發布的一項分析，大約 7% 的美國成年人（約 1800 萬人）有新冠長期症狀。哈佛大學經濟學家David Cutler 估計，到2022 年，新冠長期症狀對國家造成的總成本為3.7 兆美元，相當於新冠疫情前該國國內生產毛額(GDP) 的17%。 週三發表在《新英格蘭醫學雜誌》上的這項新研究表明，人員傷亡和經濟損失只會增加。該調查依靠電腦和先進的機器學習來篩選退伍軍人事務部維護的數百萬份醫療記錄中的數據。
---摘錄翻譯自CNN
Are cold showers good for you? What the science says
洗冷水澡對你有好處嗎？科學怎麼說
Culture
“Folks don’t just go from zero to 60 in cold showers,” he said. “They have to work up to them usually, so there is the element of your body being able to control being in a stressful environment.” This theory tracks with the results of some studies on cold showers. Participants who took showers with a water temperature of about 50 degrees Fahrenheit to 57.2 degrees Fahrenheit (10 degrees Celsius to 14 degrees Celsius) — for up to a minute daily for two weeks — reported lower stress levels than the control group, according to an October 2022 study published in the journal Current Psychology. The benefits were stronger when participants took their showers after doing a few rounds of a breathing technique that involved deep breathing, exhalation and breath holding. Simon takes a cold shower once or twice per week, which he said has helped him become more mentally present in general since the practice forces him to process the “suffering.” Cold showers may also improve mood and energy or mental alertness, which Dr. Rachelle Reed, an exercise physiologist in Athens, Georgia, has experienced, she said. “You sort of feel a little bit elated, and that is thought to be due at least in part to that increase in (the) neurotransmitters epinephrine, norepinephrine and dopamine.”
---from CNN
「人們不會只是在冷水澡中從 0 度變到 60 度，」他說。 “他們通常必須努力工作，所以你的身體有一個因素能夠控制壓力環境。” 這個理論與一些關於冷水淋浴的研究結果相吻合。根據 10 月的一份報告，在兩週內每天用 50 華氏度至 57.2 華氏度（10 攝氏度至 14 攝氏度）的水溫淋浴一分鐘的參與者報告稱，他們的壓力水平低於對照組。研究發表在2022年10月《當代心理學》期刊。當參與者在進行幾輪涉及深呼吸、呼氣和屏氣的呼吸技巧後洗澡時，效果會更明顯。 Simon每週洗一到兩次冷水澡，他說這幫助他變得更加精神集中，因為這種做法迫使他處理「痛苦」。 她說，冷水淋浴還可以改善情緒、精力或精神警覺性，喬治亞州雅典的運動生理學家Rachelle Reed博士經歷過這種情況。 “你會感到有點興奮，這被認為至少部分是由於神經遞質腎上腺素、正腎上腺素和多巴胺的增加。”
---摘錄翻譯自CNN
Two killed as walkway collapses at Italy’s notorious ‘Gomorrah’ slum
義大利臭名昭著的「蛾摩拉」貧民窟人行道倒塌，造成兩人死亡
Culture
Two people were killed and at least 13 injured Tuesday when an elevated walkway collapsed in a condemned Italian slum notorious for its links to organized criminal groups. The Italian Fire Brigade said a 29-year-old man and a 35-year-old woman were killed in the collapse at the Le Vele housing slum in the Neapolitan suburb of Scampia. More than 800 people were living as squatters in the apartment complex at the time, including 300 children, according to the fire brigade. All have been evacuated to tent camps set up by the country’s civil protection agency. The complex was made famous in Roberto Saviano’s “Gomorrah” book, film and television series, which detailed the exploits of the Camorra, a powerful Neapolitan mafia-like group. Several criminal groups tied to the Camorra operated out of the apartments and wings, many of which were protected behind heavy fencing and bullet-proof glass, according to local police who regularly raided the buildings. It was also notorious as a venue for drug deals. The housing complex, built in the 1970s and 1980s, originally consisted of four apartment buildings shaped like sails (“vele”) joined by elevated walkways. But in 2020 regional authorities ordered the site to be cleared and razed. Since then, three of the buildings have been demolished and only one now remains. The cause of the collapse at that one remaining building is not yet clear, but the structure has been condemned for years and amenities like running water, electricity and gas are all brought in illegally.
---from CNN
週二，義大利一處因與有組織的犯罪集團有聯繫而臭名昭著的貧民窟，一條高架人行道倒塌，造成兩人死亡、至少 13 人受傷。 義大利消防隊表示，那不勒斯郊區 Scampia 的 Le Vele 貧民窟倒塌事故造成一名 29 歲男子和一名 35 歲女子死亡。 據消防隊稱，當時有 800 多人非法居住在公寓大樓內，其中包括 300 名兒童。所有人都已疏散到該國民防機構設立的難民營地。 該建築群因Roberto Saviano的《蛾摩拉》(Gomorrah) 一書、電影和電視劇而聞名，其中詳細介紹了Camorra（一個強大的那不勒斯黑手黨組織）的功績。 據經常突襲搜查這些建築的當地警方稱，幾個與Camorra有關的犯罪團夥在公寓和側翼內活動，其中許多公寓和側翼都受到重型圍欄和防彈玻璃的保護。它也以毒品交易場所而臭名昭著。 該住宅區建於1970年代和1980年代，最初由四棟形狀像帆 (“vele”) 的公寓大樓組成，並通過高架走道相連。但在 2020 年，地方當局下令清理並夷平該地點。從那時起，其中三棟建築已被拆除，現在只剩下一棟。 剩下的一棟建築倒塌的原因尚不清楚，但該建築多年來一直受到譴責，自來水、電力和天然氣等設施都是非法引進的。
---摘錄翻譯自CNN
AI models fed AI-generated data quickly spew nonsense
人工智慧模型輸入人工智慧產生的資料後，很快就會吐出無稽之談
Sciences
Training artificial intelligence (AI) models on AI-generated text quickly leads to the models churning out nonsense, a study has found. This cannibalistic phenomenon, termed model collapse, could halt the improvement of large language models (LLMs) as they run out of human-derived training data and as increasing amounts of AI-generated text pervade the Internet. “The message is we have to be very careful about what ends up in our training data,” says co-author Zakhar Shumaylov, an AI researcher at the University of Cambridge, UK. Otherwise “things will always, provably, go wrong,” he says.” The team used a mathematical analysis to show that the problem of model collapse is likely to be universal, affecting all sizes of language model that use uncurated data, as well as simple image generators and other types of AI. The researchers began by using an LLM to create Wikipedia-like entries, then trained new iterations of the model on text produced by its predecessor. As the AI-generated information — known as synthetic data — polluted the training set, the model’s outputs became gibberish. The ninth iteration of the model completed a Wikipedia-style article about English church towers with a treatise on the many colours of jackrabbit tails. More subtly, the study, published in Nature on 24 July, showed that even before complete collapse, learning from AI-derived texts caused models to forget the information mentioned least frequently in their data sets as their outputs became more homogeneous.
---from Nature
一項研究發現，在人工智慧生成的文本上訓練人工智慧（AI）模型很快就會導致模型產生無意義的結果。這種被稱為模型崩潰的同類相食現象可能會阻礙大型語言模型(LLM) 的改進，因為它們耗盡了源自人類的訓練數據，並且隨著越來越多的人工智慧生成的文本遍布網際網路. 「我們傳達的訊息是，我們必須非常小心訓練資料中的最終內容，」共同作者、英國劍橋大學人工智慧研究員 Zakhar Shumaylov 說。否則，「事實證明，事情總是會出錯」，他說。該團隊透過數學分析表明，模型崩潰問題可能是普遍存在的，會影響使用未經整理數據的所有規模的語言模型，以及簡單的圖像生成器和其他類型的人工智慧。 研究人員首先使用一個大型語言模型(LLM)創建類似維基百科的條目，然後將新的模型迭代訓練在由其前身生成的文本上。由於人工智慧產生的資訊（稱為合成資料）污染了訓練資料庫，模型的輸出變得毫無意義。該模型的第九次迭代完成了一篇關於英國教堂塔樓的維基百科風格的文章，內容卻變成關於長耳大野兔尾巴多種顏色的論文。 更巧妙的是，7 月24 日發表在Nature 上的這項研究表明，即使在完全崩潰之前，從人工智慧衍生的文本中學習也會導致模型忘記數據集中最不常提到的信息，因為它們的輸出變得越來越同質化。
---摘錄翻譯自Nature
Exclusive: the Trump administration demoted this climate scientist — now she wants reform
獨家：川普政府降職了這位氣候科學家——現在她想要改革
Sciences
A climate scientist who was demoted for speaking out by the administration of former US president Donald Trump is seeking an investigation into her case and demanding changes to personnel policies to prevent similar retaliation against others in future. Her supporters say the proposed reforms could help her agency, the US Geological Survey (USGS), as well as others safeguard science in the event of a second Trump administration, which many fear will be even more efficient than the first at sidelining science and scientists. “This is not about what happened to me, it’s about what could happen to others,” Virginia Burkett, the scientist, told Nature, emphasizing that stronger protections are needed regardless of who wins the US presidential election in November. Burkett, whose position and salary as chief scientist for climate and land use at the USGS have been upgraded under President Joe Biden, laid out her allegations in a 200-plus-page complaint filed today with a federal watchdog agency that represents whistleblowers. She is represented by the Climate Science Legal Defense Fund, a non-profit organization based in New York. After Trump took office in 2017, Burkett repeatedly opposed what she saw as harmful actions taken by his administration, including dismantling climate research programmes, cutting science budgets and attempting to water down an influential government report on global warming. Burkett says that she was demoted from her leadership post at the USGS and then removed from a prominent White House panel that manages the government’s main, multiagency climate programme. Now she is seeking an investigation into what she calls “abuse of authority and gross mismanagement” under Trump.
--from Nature
一名因發表言論而被美國前總統Donald Trump政府降職的氣候科學家正在尋求對她的案件進行調查，並要求改變人事政策，以防止未來對其他人進行類似的報復。她的支持者表示，擬議的改革可以幫助她的機構、美國地質調查局（USGS）以及其他機構在Trump第二屆政府執政的情況下保護科學，許多人擔心Trump政府比第一次執政更加排斥科學和科學家。 科學家Virginia Burkett 告訴《自然》雜誌：「這與發生在我身上的事情無關，而是關乎其他人可能發生的事情。」她強調，無論誰贏得11 月的美國總統大選，都需要更強有力的保護。 Burkett在美國地質調查局氣候和土地利用首席科學家的職位和薪水在Joe Biden總統的領導下得到了提升，Burkett在今天向代表舉報人的聯邦監督機構提交的一份長達200 多頁的投訴中闡述了她的指控。她由總部位於紐約的非營利組織氣候科學法律辯護基金會代表。 在Trump 2017年上任後，Burkett多次反對他的政府採取的有害行動，包括取消氣候研究計畫、削減科學預算以及試圖淡化一份有影響力的政府關於全球暖化的報告。Burkett說，她被從美國地質調查局的領導職位上降職，然後又被從白宮一個負責管理政府主要的多機構氣候計畫的重要小組中除名。現在，她正在尋求對Trump領導下的「濫用職權和嚴重管理不善」的行為進行調查。
--摘錄翻譯自Nature