Komawa Labarai
SamfuraAI Understanding takaitaccen bayani

Kamfanin Frontier Enterprise ya ba da rahoton ɗakunan karatu na NUS sun ƙaddamar da AI Sense Maker

NUS Libraries da NUS IT sun ƙaddamar da AI Sense Maker, wani dandalin bincike na tattaunawa wanda ke bincika a cikin 150,000 kayan da aka ƙididdigewa kuma yana haifar da taƙaitaccen bayani, ra'ayoyi masu dangantaka, nassoshi da zane-zane na ilimi, Rahoton Kasuwancin Frontier.

6 min readRead the linked source
Source-provided image accompanying Frontier Enterprise reports NUS Libraries launched AI Sense Maker
Tushen tusheAn rubuta tushen tushe
Mawallafi
frontier-enterprise.com
Tushen hanyar haɗin gwiwa
frontier-enterprise.comhttps://www.frontier-enterprise.com/nus-libraries-launches-ai-sense-maker-a-conversational-platform-for-learning/
Nau'in tushe
Tushen da aka haɗa - ba a kafa matsayin tushen farko ba.
MaganaFahimtar wannan a cikin daƙiƙa 60

Fara a nan

Mabuɗin sharuddan

Babban Samfurin Harshe (LLM)
Samfurin harshe da aka horar akan babban haɗin gwiwar rubutu don samarwa da tantance rubutu.
Binciken Semanci
Bincika wanda ya dace da ma'ana maimakon madaidaicin maɓalli na maɓalli, galibi ta amfani da abubuwan sakawa.
AI mai ƙirƙira
Tsarin AI waɗanda ke samar da sabon abun ciki kamar rubutu, hotuna, sauti, bidiyo, ko lamba.
Gwada kankaChatGPT & LLMs Tambayoyi

Me ya faru

Kamfanin Frontier Enterprise ya ba da rahoton cewa Jami'ar Ƙasa ta Singapore ta ƙaddamar da AI Sense Maker, wani dandalin tattaunawa na yanar gizo wanda NUS Libraries da NUS Information Technology suka haɓaka. Dandalin yana amfani da babban samfurin harshe na OpenAI da bincike na ma'ana don taimakawa masu amfani su fara bincike tare da tambayoyin harshe na halitta maimakon takamaiman kalmomi na bayanai.

Kamfanin Frontier Enterprise ya ba da rahoton cewa ɗakunan karatu na NUS sun ƙaddamar da AI Sense Maker, ko ASM, tare da fasahar Watsa Labarai ta NUS. Rahoton ya bayyana ASM a matsayin dandalin tattaunawa na yanar gizo kuma ya ce NUS ita ce jami'a ta farko a kudu maso gabashin Asiya don amfani da AI ta wannan hanyar. Wannan bayanin na yanki-farko an gabatar da shi ta hanyar bugawa kuma ba a tabbatar da kansa ta hanyar tushen abin da aka bayar anan ba. An bayyana dandalin a matsayin babban samfurin harshe na OpenAI, amma rahoton bai bayyana sigar ƙirar ba, tsarin turawa ko ƙirar fasaha.

A cewar Frontier Enterprise, masu amfani za su iya farawa da tambaya a cikin yaren yau da kullun, kamar me yasa wasu unguwanni a Singapore ke jin zafi fiye da sauran. ASM yana amfani da bincike na ma'ana don fassara ma'ana da niyyar tambaya maimakon dogaro da matches na zahiri kawai. Ayyukan da aka ruwaito sun haɗa da dawo da hanyoyin da suka dace, haɓaka ra'ayoyi masu alaƙa, samar da taƙaitacciyar taƙaitacciyar jigo, ƙirƙira nassoshi don ƙarin karantawa da ƙirƙirar jadawali na ilimin mu'amala waɗanda ke nuna alaƙa tsakanin ra'ayoyi. Waɗannan su ne bayanin iyawar dandali da aka ruwaito, ba sakamakon da aka auna kansa ba.

Kamfanin Frontier Enterprise ya ba da rahoton cewa ASM da farko yana zana abubuwa 150,000 na dijital. Tarin ya haɗa da takaddun bincike daga ScholarBank, ma'ajiyar bincike ta NUS, da kayan aiki daga Digital Gems, ƙoƙarin ƙididdige ɗakunan Laburaren NUS da aka mayar da hankali kan tarin musamman da tushen farko, musamman waɗanda ke da alaƙa da kudu maso gabashin Asiya. Labarin ya ce waɗannan tarin sun haɗa da littattafan da suka kasance a ƙarni na 17. Ba ya ƙayyadadden adadin abubuwan da ake iya bincikawa gabaɗaya, yadda aka jera kafofin, ko dandamali zai iya samun damar yin amfani da sikanin asali, ko duk masu amfani suna karɓar ɗaukar hoto iri ɗaya.

Rahoton ya ambato Natalie Pang, mataimakiyar farfesa kuma ma’aikaciyar laburare a jami’a a NUS, tana mai cewa masu binciken ba su da lokaci da tsafta kuma ana nufin ASM ne don taimaka wa mutane su fahimci bayanan maimakon samun su kawai. Pang ya kuma ce tsarin yana nufin tallafawa mafi kyawun tambayoyi da bincike mai zurfi maimakon maye gurbin tunani mai mahimmanci. Waɗannan maganganun ana danganta su zuwa Pang ta hanyar Kamfanin Frontier; Madogararsa ba ta ba da gwajin mai amfani mai zaman kansa ba, nazarin kwatance ko ƙima na ko kayan aikin ya cimma waɗannan manufofin.

Bayanan tushe: frontier-enterprise.com ↗

Me ya sa yake da mahimmanci

Aiki yana nuna yadda jami'a ke sanya AI mai haɓakawa kai tsaye cikin binciken bincike da tarin ɗakin karatu. Rahoton da aka ruwaito na farko ya haɗu da takaddun bincike tare da ƙayyadaddun kayan da ba kasafai aka ƙididdige su ba da tushen farko na Kudu maso Gabashin Asiya, mai yuwuwar baiwa masu amfani damar samun hanyar samun haɗin kai a cikin tarin. Labarin baya tabbatar da daidaito, samuwa, aiki ko tasiri akan sakamakon koyo da kansa.

Ƙaddamar da rahoton da aka ruwaito saboda yana sanya AI a cikin wani muhimmin mataki na aikin ilimi: yanke shawarar inda bincike ya fara. Binciken laburare na al'ada yakan buƙaci masu amfani su san wace rumbun adana bayanai don bincika da kuma waɗanne kalmomi za su yi amfani da su. Keɓancewar zance na iya rage wannan shingen ta barin a bayyana tambaya cikin yare na yau da kullun. Maido da ma'anar ma'ana zai iya fitowa fili wanda ke amfani da ƙamus daban-daban daga kalmomin mai amfani. Kasuwancin Frontier yana gabatar da waɗannan azaman fa'idodin da aka yi niyya, amma baya samar da ingantaccen ma'auni a cikin abin tunawa, dacewa ko lokacin bincike.

Dabarun tattarawa yana ba da aikin tura mahimanci fiye da babban abin tattaunawa. Ta hanyar haɗa ma'ajiyar takaddun bincike tare da tarawa na musamman da aka ƙirƙira, an ba da rahoton ASM don haɗa guraben karatu na zamani tare da maɓuɓɓugan tarihi. Idan tsarin yana adana hanyoyin haɗin yanar gizo kuma ya bayyana gaskiyar amsoshi a sarari, zai iya taimaka wa ɗalibai da masu bincike su gano kayan da ke da wahalar ganowa ta hanyar bincike daban-daban. Ƙimar ƙimar jama'a ta fi dacewa musamman ga tarihin kudu maso gabashin Asiya da malanta, inda ƙididdigewa da ganowa na iya shafar wanda zai iya shiga tare da tushen farko.

A lokaci guda, babban tarin maƙasudin ba ya kafa ingantaccen taimako na bincike da kansa. Takaitattun abubuwan da ƙirar harshe ke samarwa na iya ƙetare cancantar cancanta, daidaita rashin jituwa ko gabatar da da'awar waɗanda kayan da aka dawo ba su da tallafi. Binciken Semanci kuma na iya dawo da tushe iri ɗaya na ra'ayi waɗanda basu dace da wata tambaya ba. Majiyar ba ta faɗi ko ASM tana nuna ƙididdiga ba, ambaton matakin shafi, alamun tabbaci, gibin tarin ko faɗakarwa lokacin da shaida ta yi rauni. Hakanan baya bayar da rahoton kimantawa akan masu karatu, bincike na al'ada ko wasu kayan aikin bincike.

Don haka aikin yana wakiltar gwaji mai amfani na cibiyoyi, amma har yanzu tasirinsa bai auna ba a cikin rahoton da aka kawo. Babu bayani game da adadin masu amfani, ƙuntatawa damar shiga, farashin aiki, tanadin samun dama, shirye-shiryen lasisi ko sakamakon koyo. Rahoton kuma bai tabbatar da ko tsarin yana samuwa fiye da al'ummar NUS ba. Shaida mafi ƙarfi ita ce ɗakin karatu na jami'a ya ƙaddamar da wani dandamali mai suna AI tare da ƙayyadaddun tarin farko da ayyuka da yawa da aka bayyana; iƙirari mai faɗi game da canza ilimi ko bincike zai buƙaci ƙarin shaida.

Interactive Mechanism

Ingantacciyar hanyar sadarwa: Yadda A zahiri yake Aiki

Bincika fasahar da ke bayan wannan ci gaban ta hanyar mu'amala.

Thinking Budget (Test-Time Tokens):1,024 tokens
Complex Accuracy79%Math & Code Logic
Latency3.2sTime to first full output
Inference Cost$0.0092Per query estimated
Reasoning StyleStep VerificationInternal chain depth
Active Thinking Trace:
1Deconstruct user problem into formal constraints
2Propose candidate hypotheses & step-by-step calculation
3Self-correction: Backtrack and refute subtle edge cases
4Exhaustive consistency check & final output synthesis
Core takeaway: Test-time compute fundamentally changes AI economics. Instead of only scaling during pre-training, giving reasoning models more tokens at inference time allows them to systematically solve PhD-level STEM problems.
Duba ra'ayi na hulɗa+10 Points
ChatGPT & LLMs Quiz

What is a common training objective for an autoregressive language model?

Abin kallo na gaba

Tambayoyi masu mahimmanci na gaba sune yadda AI Sense Maker ke ƙididdigewa kuma ya bambanta maɓuɓɓuka, yadda yake magance kurakurai a cikin kayan tarihi, wanda zai iya samun damar yin amfani da shi, da kuma ko tushe na 150,000 zai fadada. Masu amfani da cibiyoyi kuma yakamata su kalli shaida game da keɓantawa, haƙƙin mallaka, ɗaukar hoto, bita na ɗan adam da ko kayan aikin yana inganta ingancin bincike maimakon kawai rage lokacin bincike.

Na farko, masu sa ido su nemi takamaiman bayani game da isa da iyaka. Rahoton bai ce ko AI Sense Maker yana buɗe wa jama'a ba, iyakance ga ɗaliban NUS da masana ilimi, ana samun su ta hanyar matukin jirgi kawai ko kuma an riga an tura shi gabaɗaya. Hakanan baya bayyana ko masu amfani zasu iya bincika kowane abu a cikin tushe na kayan 150,000 ko kuma samun dama ya bambanta bisa ga haƙƙin mallaka, sirri, tsarin tarin ko matsayi na digitization. Waɗannan cikakkun bayanai za su ƙayyade ko ƙaddamarwa da farko sabis ne na harabar ko kuma babban tushen ilimin jama'a.

Na biyu, ingancin sarrafa tushen zai zama tsakiya. Tsarin bincike mai fa'ida ya kamata ya bayyana waɗanne takardu ne ke goyan bayan amsa, bambanta taƙaitaccen abin da aka samar ta atomatik daga shaidar da ke ƙasa da kuma samar da hanya mai amfani don bincika ainihin abu. Tambayoyi masu mahimmanci sun haɗa da ko nassoshi daidai ne, ko an ƙirƙiri jadawali na ilimi daga ƙayyadaddun alaƙa ko ƙirar ƙira, yadda ake sarrafa kwafi da maɓuɓɓuka masu karo da juna, da abin da zai faru idan tambaya ta faɗi a wajen tarin. Kasuwancin Frontier baya bayar da rahoton waɗannan kariyar ko kowane gwajin daidaito.

Na uku, zaɓin shugabancin jami'a ya cancanci kulawa saboda tsarin ya ƙunshi takaddun bincike na zamani da kayan tarihi da ba safai ba. Ya kamata rahoto na gaba ya fayyace yadda ake adana tambayoyin mai amfani, ko ana amfani da mu'amala don haɓaka ƙirar ƙira, yadda ake kiyaye tarin ƙima ko ƙuntatawa, da kuma yadda ake sarrafa haƙƙin mallaka da lamuni. Majiyar kuma ba ta gano ƙirar OpenAI ba ko faɗi ko NUS tana aiki da ƙarin tacewa, dawo da ko sake duba yadudduka kewaye da shi.

A ƙarshe, ƙima mai ma'ana zai nuna ko ASM yana inganta bincike maimakon kawai sanya hulɗar farko ta sami sauƙi. Shaidu masu fa'ida na iya haɗawa da kwatancen sarrafawa tare da binciken ɗakin karatu na yanzu, gwaje-gwajen da suka haɗa da kalmomin da ba a sani ba da tarin tarihi, nazarin kurakurai daga masu ɗakin karatu, da martani daga ɗalibai da masu bincike kan lokaci. Fadada tarin, gyare-gyaren da aka rubuta da kuma shaidar ci gaba da amfani za su nuna ko ƙaddamarwar ta zama kayan more rayuwa mai dorewa. Idan ba tare da irin wannan bayanin ba, tasirin aiki ya kasance mai ban sha'awa amma ba a tantance ba.

Jagorori masu alaƙa & tambayoyin tambayoyi

ChatGPT da LLMAI Model ya bayyanaƊa'a ta AIAI horoGwada abin da kuka sani - gwada gwajin AI kyautaNemo kalmar AI a cikin ƙamus ɗin muBi samfurin AI na sakin tracker
An sami wannan yana da amfani?