Fitowar 16 Oktoba, 2026

Taurari suna zuwa da launin lemu, fari ko shuɗi, gwargwadon zafinsu. Fararen dodanni (white dwarfs), ragowar matattun taurari masu matuƙar cunkoso, suna karya wannan doka. Wata ƙungiyar Birtaniya da Jamus ta lissafa ainihin launin da za su nuna daga kusa.

Hanyarsu: mayar da hasken kowane tauraro zuwa launin da idon ɗan adam zai gani, kamar yadda allon kwamfuta ke yi. Wasu sararin iskokinsu suna da cunkoso har carbon, calcium ko sodium suna haɗiye dukkan ratsin launi.

Sakamakon wani tarin launuka ne. Taurarin carbon suna zama ruwan hoda ko shunayya. Wani tauraro, WD J1330+6435, zai yi kama da kore mai haske (chartreuse): ba a san wani nau'in tauraro da yake kore ba. Wani kuma, a digiri 4,000 kawai, shuɗi ne mai haske.

Duk da haka, a cikin fararen dodanni 152 da ke tsakanin parsec 20 daga Rana, launin tsakiya (median) yana kusa da fari zalla. Kusan 3% ne kawai ke kauce masa nesa. Sunan ya dace bayan komai.

Tushe: Any colour you like: The diverse chromaticities of white dwarfs, https://arxiv.org/abs/2610.18966

Piet Mondrian ya zana zane-zane da aka yi da rectangle masu launi. Masana lissafi sun mayar da su wasanin hankali. Yanka murabba'i zuwa rectangle masu gefuna na lambobi cikakku, dukansu daban. Yaya filayensu za su iya kusantar juna?

Giɓin da ke tsakanin mafi girman fili da mafi ƙanƙanta ana kiransa naƙasu (defect). Naƙasu na sifili zai zama yanka cikakke. Babu wanda ya taɓa samun ɗaya, kuma galibi sun yarda cewa giɓin dole ne ya ci gaba da girma tare da murabba'in.

Masana lissafi biyu na Florida sun tabbatar da akasin haka. Fara da murabba'i mai gefen 315 da aka yanka zuwa rectangle bakwai: filaye daga 14,000 zuwa 14,280. Sannan a gina murabba'ai mafi girma, faɗin 106,785 da sama da haka, har abada.

Kowanne yana riƙe da giɓi iri ɗaya: 280. Dabarar ita ce wata lissafin ka'idar lambobi, ta Pell, da aka yi amfani da ita kan √19. Har yanzu ba a kawar da yanka cikakke ba, kuma har yanzu ba a same shi ba.

Tushe: A Bounded-Defect Construction for the Mondrian Art Problem, https://arxiv.org/abs/2610.19193

Mai caca da yake yin caca ta adalci kawai ba zai iya tsammanin cin nasara ba. Damar ninka kuɗinsa sau 1,000 kafin ya yi fatara ita ce 1 cikin 1,000 a mafi yawa. Wannan shi ne sanannen "faɗuwar mai caca" (gambler's ruin).

Yanzu a ba da damar karɓar bashi, da caca har zuwa ninki 10 na abin da mai caca ya mallaka. Wani masanin kimiyyar kwamfuta na Carnegie Mellon ya gano mafi kyawun dabara. Ci ɗan kaɗan a kowace caca, rasa komai lokaci-lokaci kawai.

Sakamakon: kusan damar 50% ta mayar da $1 zuwa $1,000 ba tare da taɓa yin fatara ba. A gaba ɗaya, damar ninka kuɗinka sau r kusan r ne a ƙarfin debe 1 bisa leverage.

Lehman Brothers ta yi aiki da leverage na 20 zuwa 40. Da caca ta adalci, girma ninki takwas yana da damar 90%. Lokacin da ta faɗi, darajarta ta kai kusan debe dala biliyan 150. Wani yana faɗuwa, sai dai ba mai caca ba.

Tushe: With Leverage, the Gambler is Not the One Who Gets Ruined, https://arxiv.org/abs/2610.19239

Ranar 10 ga Mayu 2024, gajimaren plasma na rana sun bugi Duniya. Ita ce guguwar maganaɗisun duniya mafi ƙarfi cikin sama da shekaru 20. An faɗakar da miliyoyin mutane. Da yawa sun fita waje, wayoyi a hannu.

Galibin na'urorin kimiyya sun yi nisa a arewa, ko kuma dararen bazara masu haske sun makantar da su. Don haka masu bincike suka tambayi jama'a. Sun tattara rahotanni 696 daga ƙasashe sama da 30, da ƙarin 186 daga wani kundin ƙasar Finland.

An ga aurora galibi tsakanin digiri 40 da 60 na latitud ɗin maganaɗisun duniya, wasu rahotanni ma ƙasa da haka. Wannan ya fi kusa da ma'aunin ƙasa (equator) sosai fiye da yadda samfura suka yi hasashe. An ga launin kore da ido tsirara har zuwa kusan digiri 38.

Galibin shaidu sun ga ja da ruwan hoda, masu haske isa ido ya gani. Wannan yana nuna ambaliyar electron masu ƙaramin kuzari. A gaba, ingantaccen fom zai iya mayar da kowace waya kayan aunawa.

Tushe: The Gannon Storm: Citizen science observations during the geomagnetic superstorm of 10 May 2024, https://arxiv.org/abs/2610.19929

Ion da aka tarko suna cikin mafi kyawun tubalan gina kwamfutocin kwantum. Filayen lantarki suna riƙe su suna shawagi a cikin vacuum. Amma waɗannan na'urori suna aiki ne kawai a tsaye, a kan teburin dakin gwaje-gwaje.

Wata ƙungiya daga Innsbruck da Siegen ta saka cikakken tarko a cikin firam mai tsayin mita 1.8. Sun ɗora shi a mota, suka ba shi wuta da na'urorin wutar ajiya guda biyu, suka hau babbar hanya.

Tafiya ta farko: kilomita 18.6 kaɗan bayan iyakar Austriya da Italiya, mintuna 15, kusan ion 3,800 har yanzu a tarko lokacin isowa. Tafiya ta biyu: kilomita 63.4 a Jamus, mintuna 45, kusan ion 1,170 suka rage. Wannan asarar ta yi daidai da abin da tarkon yake rasawa a tsaye a dakin gwaje-gwaje.

Wannan ita ce tafiya ta farko irin ta ga wannan nau'in tarko, wanda galibin dakunan gwajin kwantum suke amfani da shi. An ajiye ion ɗin ne kawai, ba a sanyaya su ko a yi amfani da su a matsayin qubit ba tukuna. Wannan shi ne mataki na gaba, a kan tafiya.

Tushe: Demonstration of a mobile ion-trap quantum device, https://arxiv.org/abs/2610.20487

Me ya sa cukun tsirrai ba ya taɓa jin kamar na gaske? Gwajin yanayin taɓawa (texture) yawanci yana nufin murƙushe samfurin. Wata ƙungiya a Leeds ta gwada sauraron sa maimakon haka.

Sun tura ultrasound ta cikin iska da ta cikin yankuna 119 na cuku iri 13 na manyan shaguna, tara na madara da huɗu na tsirrai. Bincike ɗaya yana ɗaukar kusan minti ɗaya kuma yankan yana nan lafiya.

Taswirorin suna bambanta su a kallo ɗaya. Tsohon Cheddar yana nuna faci masu faɗin milimita, waɗanda dunƙulen curd ɗinsa da suka haɗe suka bari; yankunan tsirrai kuma iri ɗaya ne ko'ina. A matsakaici, dukansu suna da ƙarfi daidai. Amma ƙarfin cukun madara yana kaiwa ninki 7.5, na tsirrai kuma ninki 1.9 kawai.

Sannan wata ƙaramar cibiyar jijiyoyi ta wucin gadi (neural network) tana hasashen cikakken lanƙwasan matsewa na kowane yanki daga binciken kawai. Yankunan tsirrai uku suna aiki kamar Cheddar, ɗaya kamar Emmental, babu kamar mozzarella. Wannan kusurwa mai laushi har yanzu babu kowa.

Tushe: Sounds Cheesy - Bridging the Texture Gap Between Dairy and Plant-Based cheese with Non-Contact Ultrasound and AI, https://arxiv.org/abs/2610.19924

Ana matse kiɗan da ake yaɗawa (streaming): ana jefar da wani ɓangare na sautin, inda ya kamata kunne ba zai lura ba. Amma daga wane inganci ne ba a iya jin bambancin da gaske? Masu bincike biyu na Deezer sun gwada Opus, wani codec na buɗaɗɗen tushe.

Gwajin ya yi tsauri: makafi biyu, a cikin ɗakuna marasa shigar sauti, da belun kunne masu rufe. Masu sauraro sun ji asalin, sannan nau'i biyu don kwatantawa: ɗaya da aka matse, ɗaya kuma a ɓoye daidai yake da asalin.

Da masu sauraro 21 da gutsuttsura 23 masu wahala, kilobit 128 a daƙiƙa ya wuce. Sannan ƙungiyar ta riƙe kunnuwa 10 mafi kaifi da gutsuttsura 8 mafi wahala kawai. A ƙarƙashin wannan bincike, an kama 128.

A kilobit 160 a daƙiƙa, waɗannan kunnuwan ba su sake bambance shi da CD ba, kuma ƙididdiga ta tabbatar da haka. Wasu kayan aiki da ƙarin ƙwararru za su sa hukuncin ya zama na duniya baki ɗaya.

Tushe: Evaluating perceptual transparency of lossy music audio compression with Opus codec, https://arxiv.org/abs/2610.20246

Wani ra'ayi game da tsufa: lalacewa ta bazata tana taruwa a kowace ƙwayar halitta, maye gurbi bayan maye gurbi. Sarrafa kwayoyin gado yana warwarewa a hankali. Amma kowace ƙwaya tana lalacewa ta hanyarta. Ta yaya za mu ga tsari ɗaya na bai ɗaya?

Wata ƙungiya daga Isra'ila, China da Amurka ta horar da cibiyar jijiyoyi ta wucin gadi (neural network) kan ayyukan kwayoyin gado a cikin ƙwayoyin halitta ɗaiɗai. Tana iya kashe kwayar gado ɗaya ta kwamfuta, a cikin kowace ƙwaya, ta kuma hasashen illolin da za su biyo baya.

A cikin nau'ikan ƙwayoyin halitta 45 na mutane, ɓeraye, ƙudaje da tsutsotsi, tasirin kashe kwayar gado ɗaya ya fi bambanta tsakanin tsofaffin ƙwayoyi a cikin 41. Ƙwayoyin da ke da ƙarin maye gurbi, ko ƙarin alamun tsufan ƙwaya (senescence), su ne suka fi kaucewa.

Da alama tsofaffin ƙwayoyi suna rasa ƙa'idojinsu na bai ɗaya. Waɗannan hasashe ne na samfurin kwamfuta, ba kashe kwayar gado a dakin gwaje-gwaje ba, kuma alaƙa ce ba sanadi ba. Gwaje-gwaje na gaske za su tabbatar da su.

Tushe: Deep learning-based virtual perturbation reveals increasing heterogeneity of gene regulatory responses with aging, https://arxiv.org/abs/2610.19029