Fitowar 29 Oktoba, 2026

3I/ATLAS shi ne abu na uku kawai da aka taɓa gani yana zuwa daga wani tauraro. A Disamba 2025, na'urar hangen nesa James Webb ta yi nazarin ƙurar da ke kewaye da cibiyarsa ta ƙanƙara.

Ƙura mai ɗumi tana haskakawa a hasken infrared. Silicates, ma'adinan dutse, suna barin kumburi kusa da micrometre 10. A taurarinmu masu wutsiya (comets), wannan kumburi yana yawan ɗauke da alamun lu'ulu'u. A nan kuwa, santsi ne.

Mafi kyawun dacewa: lu'ulu'u sifili, 5.3% a mafi yawa. Ƙurar amorphous ce, ba ta da tsari kamar gilashi. A taurari masu wutsiya, lu'ulu'u suna tona asirin ƙurar da ta taɓa yin zafi. Wannan ƙurar da ƙyar ta yi.

Wasu ƙungiyoyi sun sami alamun haihuwa ƙasa da kelvin 30, daga iskar gas ɗinsa. Ƙurar ta yarda. Mai yiwuwa wannan tauraro mai wutsiya ya girma a wani lungu mai sanyi da natsuwa na faifan (disc) gidansa. Ƙaramin cakuɗuwa, kusan babu zafi.

Tushe: JWST/MIRI Observations of Amorphous Silicates in Interstellar Comet 3I/ATLAS, https://arxiv.org/abs/2610.37540

Wata wuta tana yaɗuwa a fili da gudu ɗaya ta kowace hanya. Kana gina bango kewaye da ita a lokacin da abin ke faruwa (real time), bulo bayan bulo. Da wane gudu ne dole ka gina don ka kulle ta?

Masanin lissafi Alberto Bressan ya daɗe yana hasashen amsar: fiye da ninki biyu na gudun wutar. Har yanzu babu wanda ya tabbatar da hakan, ko ya sami hanyar kauce masa.

Tare da Luca Talamini, yanzu ya nuna cewa ninki 1.626 na gudun wutar ba ya taɓa isa. Iyakar da ta gabata ita ce 1.5. Hujjar tana kwatanta ƙasar da ta ƙone da kumfar sabulu guda biyu.

Ga bango da aka rufe a da'ira guda ɗaya, sun warware shi. Sama da 2, ko kuma wutar ta kuɓuta. Dabara mafi arha za ta buƙaci bangaye masu ban mamaki. Aikinsu kawai: rage gudun wutar, kafin ta haɗiye su.

Tushe: On the Minimum Speed for a Fire Blocking Problem, https://arxiv.org/abs/2610.36919

Wani karo da aka yi hasashensa a 1974: neutrino yana billa daga dukkan cibiyar zarra lokaci ɗaya. Saboda dukkan cibiyar tana shiga ciki, na'urorin ganowa (detectors) na 'yan kilogram za su iya kama shi.

CONUS+ yana a mita 20.7 daga zuciyar wani reactor na nukiliya a Switzerland, a Leibstadt. Lu'ulu'unsa (crystals) na germanium masu nauyin kilogram 5.6, a cikin garkuwa mai nauyi, suna jiran waɗannan ƙananan bugu.

Daga Fabrairu 2025 zuwa Yuli 2026, ƙungiyar ta ƙirga irin waɗannan abubuwa (events) 1,555. Ka'ida ta yi hasashen 1,701, a cikin iyakokin kuskure. Siginar ta fi rashin tabbacinta girma da ninki 10.2, bayan nazarin rufe ido (blind analysis).

Wannan shi ne ma'auni mafi daidaito na wannan karo kusa da reactor. Yana kuma tantance wani ma'auni (parameter) na ƙarfi mai rauni (weak force). Ƙarin shekara ɗaya na bayanai zai rage kuskuren ƙididdiga zuwa kusan 5%.

Tushe: High significance measurement of reactor coherent elastic neutrino-nucleus scattering on germanium by CONUS+, https://arxiv.org/abs/2610.37318

Ƙananan duniyoyi da ke wajen tsarin rana (exoplanets) suna taruwa a kewayen girma biyu: kusan ninki 1.3 da 2.4 na faɗin Duniya. Ana tunanin manyan suna riƙe da sararin iska mai kauri na hydrogen. Ana tunanin ƙananan sun rasa shi.

Wani shahararren ra'ayi yana ɗora laifin a kan manyan karo a ƙarshen samuwar duniyoyi. Karon yana narkar da dutsen, kuma saman da ke ƙuna yana korar iskar gas ɗin. An gama magana?

Masu bincike biyu, a Shanghai da Yale, sun sake lissafa cikakken kasafin makamashin karo. Dukkan nauyin jikin da ya yi karon (impactor) yana ƙara ƙarfin janyewa. Amma tsakiyarsa ta ƙarfe tana nutsewa, narkewa tana shanye zafi, kuma saman sararin iskar da ƙyar yake ɗumama.

A cikin duniyoyi miliyan 100 da aka kwaikwaya, karo galibi yana taimaka wa duniyoyi su riƙe hydrogen ɗinsu. Don haka ƙananan Neptune (sub-Neptunes) za su iya ɓoye cikin dutse mafi nauyi fiye da manyan Duniya (super-Earths). Na'urorin hangen nesa na gaba za su iya tabbatarwa ta hanyar auna nauyinsu da binciken sararin iskarsu.

Tushe: Retention of primordial atmospheres through giant impacts, https://arxiv.org/abs/2610.36446

Takalmin ƙugu na robot (robotic hip braces) za su iya rage wa tsokokin jiki makamashin da tafiya ke ci. Amma daidaita ɗaya yawanci yana ɗaukar dogayen zama da kowane sabon mai sanyawa.

Wata ƙungiya a New Jersey ta horar da ɗaya a cikin kwamfuta maimakon haka. Abokin aikinsa jiki ne na kwamfuta mai tsoka 304, masu girma da salon tafiya da aka zaɓa ba bisa tsari ba. Manufar: sa waɗannan tsokoki su ƙona ƙarancin makamashi.

Sannan manya takwas na zahiri sun yi tafiya da shi a kan injin tafiya (treadmill), ba tare da wani daidaitawa ko kaɗan ba. Na'urar nazarin numfashi ta bi diddigin ƙoƙarinsu. Idan aka kwatanta da sanya shi a kashe, ya rage amfanin makamashinsu da 18.3%. Kowane mai tafiya ya amfana.

Matsalar: na'urar kanta mai nauyin kilogram 5.9 tana ƙara 13.1%. Masu tafiya shida har yanzu sun kashe ƙasa da yadda suke yi ba tare da ita ba, har zuwa 24% ƙasa. Firam mai sauƙin nauyi ya kamata ya ƙara faɗaɗa wannan tazara.

Tushe: Multi-Agent Reinforcement Learning of Co-Adaptive Hip Exoskeleton Control (MARL-Exo) for Metabolic Energy Reduction, https://arxiv.org/abs/2610.37441

Ka tambayi chatbot lamba ta bazata (random) tsakanin 1 da 30. Nan take yake amsawa. Mutane suna amfani da irin waɗannan amsoshi don raba kunnen doki ko jefa tsabar kuɗi ta kama-da-wane (virtual). Amma yaya bazatar su take?

Wani mai bincike ya lissafa ainihin yiwuwar da kowane samfuri ke ba kowace lamba. Ba ɗebo amsoshi (sampling), ba hasashe. Iyalai goma sha takwas na samfurori a buɗe (open models), kewayo biyar, kwamfutoci biyu daban-daban.

Daidaitawa (tuning) tana mayar da ɗanyen samfuri mataimakin hira (chat assistant). A dukkan nau'i-nau'i 15 da aka gwada, ta sa amsoshin su rasa daidaito. Wani mataimaki yana amsa 17 kusan a kowane lokaci. Sigarsa da ba a daidaita ba ba ta ba 17 wani fifiko na musamman ba.

Samfuran da aka daidaita kusan ba sa taɓa faɗin 1 ko lamba mafi girma. Suna guje wa ninkin 10 (multiples of 10), kuma da yawa suna karkata ga lambobi mara cika (odd). Gyaran mai sauƙi ne. A bar na'urar samar da lambobin bazata ta gaske ta zaɓa, sannan a miƙa sakamakonta ga samfurin.

Tushe: 3, 7, 14, 17, 42: Instruction Tuning Installs and Sharpens Number-Choice Heuristics in Language Models, https://arxiv.org/abs/2610.36178

Ɗauki ayyukan Amurka da suka fi fuskantar AI. A can, kowace shekarar karatu tana ƙara kusan 7.5% ga albashi a da. Tun zuwan ChatGPT, tana ƙara kusan 6.9%.

Masana tattalin arziki uku a Buenos Aires sun kwatanta binciken albashi na Amurka daga 2018 zuwa tsakiyar 2026. Fitowar ChatGPT ta raba lokacin gida biyu. Sun kwatanta ayyukan da ke fuskantar AI, kamar masu haɓaka manhajoji (software developers) da akantoci, da direbobin manyan motoci ko masu hidima a gidan abinci.

Ra'ayinsu: AI tana raba ilimin da aka rubuta, don haka abin da makaranta ke koyarwa yana daina yin ƙaranci. Gwanintar da aka koya ta hanyar aikatawa ba ta zuwa da sauƙi haka daga na'ura.

Hakika, ƙwarewar farko yanzu tana ɗan ƙara biya a waɗannan ayyuka. Amma bayan kusan shekaru 18, kowace ƙarin shekara tana rage biya. Wasu ƙwarewa kamar suna da alaƙa da tsoffin hanyoyin aiki. Marubutan suna kiransa sake farashi (repricing), ba hasashen gaba ba.

Tushe: Practice Makes Perfect: Tacit Knowledge, Generative AI, and Returns to Schooling and Experience, https://arxiv.org/abs/2610.36008

Ka yanke igiya, sai bakinta ya warware. Bayanin gargajiya yana buƙatar abubuwa uku. Murɗi, jan igiyar, da gogayya tsakanin zarurrukan.

Wata ƙungiya a University of Massachusetts ta yi amfani da zarurruka da ke naɗe a yanayinsu na hutu, kamar spring. Idan aka murɗa su tare, sannan aka yanke, damin yana nan a kulle. Babu wani kube da ke riƙe shi.

A cikin kwaikwayonsu (simulations), an kusan kawar da gogayya gaba ɗaya. Ga siffofin da suka dace, damin yana nan a riƙe, ba tare da wani abu da zai kama ba. Ana matse kowane zare daga siffar da ya fi so, sai ya tunkuɗe maƙwabtansa.

Amma idan naɗin ya yi kaɗan, damin yana buɗewa daga ƙarshensa, kamar tsagewa mai yaɗuwa. Wasu siffofi, har da zarurruka madaidaita, babu yawan gogayya da zai iya ceton su. Injiniyoyi za su iya amfani da wannan iyaka wajen ƙera zarurruka da ke kwance kansu.

Tushe: No friction required: Self-locking in entangled assemblies of curved filaments, https://arxiv.org/abs/2610.36330

Fashewar duwatsu masu aman wuta, guguwar ƙura, injinan niƙa kofi. Ƙwayoyin da suka yi karo da juna suna samun cajin wutar lantarki. Har yanzu masana kimiyyar lissafi ba su iya bayyana dalilin gaba ɗaya ba. Abin da ya fi ban mamaki ma, ƙwayoyi biyu na abu iri ɗaya za su iya ƙarewa da caji masu saɓani.

Wata ƙungiya a KAIST, a Koriya ta Kudu, tana nuna zafin bugun a matsayin dalili. Sun ɗauka cewa ƙwayoyin biyu suna karɓar zafi daidai. Ƙaramar tana ƙara ɗumama fiye da babbar: tana da ƙarancin abu da zai shanye wannan zafin.

Wannan bambancin zafi, 'yan kashi cikin dubu na digiri, yana haifar da ƙarfin lantarki (voltage). Wannan shi ne tasirin Seebeck (Seebeck effect), ƙa'idar da thermocouple ke aiki da ita. Sai caji ya kwarara daga wannan ƙwaya zuwa waccan.

Abin da aka yi ƙwayoyin da shi ne ke ƙayyade alƙibla. A lissafinsu, ƙananan ƙwayoyin quartz suna zama korau (negative), manya kuma tabbatacce (positive). Da nylon, abin yana juyawa, ƙananan ƙwayoyi suna zama tabbatacce. Gwaje-gwaje sun riga sun ba da rahoton halayen biyu.

Tushe: Tribocharging of Granular Materials Dictated by Geometric and Thermoelectric Effects, https://arxiv.org/abs/2610.36757

Ƙananan lu'ulu'u (crystals) za su iya kama haske tsakanin fuskokinsu, kamar tsakanin madubai biyu. Hakan ya sa ake ganin za su iya zama laser na ƙanƙanta matuƙa. Amma tsagewa guda ɗaya kawai, sai hasken da aka kama ya ɓace.

A University of Hyderabad, a Indiya, masana kimiyyar sinadarai sun tsaga irin waɗannan lu'ulu'u masu haske ruwan lemu da gangan. Kayan aikinsu shi ne bakin na'urar hangen ƙananan abubuwa ta ƙarfin zarra (atomic force microscope), wanda aka yi amfani da shi kamar ƙaramar guduma.

Sannan suka tura gutsuttsuran su sake taɓa juna. Tsagewar ta ɓace. Har gutsure biyu da suka rabu ma sun sake haɗewa suka zama lu'ulu'u ɗaya. A cikin lu'ulu'u shida masu jagorantar haske, tsakanin 32% zuwa 99% na hasken ya dawo.

A cikin lu'ulu'u mai faɗin micrometre 31, resonance ɗin hasken sun ɓace a wurin tsagewar. Bayan warkewa, sun dawo, daidai da na lu'ulu'un da bai lalace ba. Amma yankin da aka gyara ya fi laushi: taurinsa (stiffness) ya sauka zuwa kusan kashi biyu bisa uku.

Tushe: Fracture Mechanics and Self-Healing in Molecular Crystal Resonators for Lasing, https://arxiv.org/abs/2610.36692