Chii chaitika
IT Brief Australia inoshuma kuti RevEng.AI yakatanga Mega Bite, yebhinari-yekuongororwa yekuwedzera kwepuratifomu yayo yeBinNet. Iko kuburitswa kunosanganisira WilBERT, iyo inoratidza yakafanana kodhi mukati mebinary artefacts, uye Ventris, iyo inotsanangura executable uye kuyedza kudzoreredza semantically yakaenzana sosi kodhi. RevEng.AI inoti mamodheru akadzidziswa pane anopfuura 50 matiriyoni ekodhi-to-binary tokens uye anomhanya anenge zviuru gumi tokens pasekondi. Chirevo hachibvumire chakazvimirira izvo zvichemo kana kunyora kuwanikwa kweveruzhinji, mitengo, kana rezinesi.
IT Brief Australia inoshuma kuti RevEng.AI yakatanga Mega Bite sekuwedzera kweiyo BinNet tekinoroji. Iyo pasuru ine maviri evaridzi ekugadzira-hungwaru modhi: WilBERT, yakagadzirirwa kuona yakafanana kodhi mukati mebinary artefacts, uye Ventris, yakagadzirirwa kutsanangura zvinogoneka uye kudzoreredza semantically yakaenzana sosi kodhi. Iwo mamodheru anoitirwa kuunganidzwa software, kusanganisira yakabviswa mabhinari uko yekutanga budiriro mamiriro haasisiri kuoneka.
Sekureva kweIT Brief Australia, RevEng.AI inoti iwo mamodheru akadzidziswa pane anopfuura 50 tiririyoni tokens kubva kodhi-to-binary pairs akagadzirwa nenzira dzakasiyana-siyana dzekugadzira software. Kambani iyi zvakare inoti Mega Bite inoshanda nekumhanya kwekumhanya kweanosvika zviuru gumi tokeni pasekondi. Chirevo chinotsanangura zvinogona kushandiswa mukuongorora malware, mhinduro yezviitiko, vimbiso yesoftware, uye kuongororwa kweanopa kana masystem anogarwa nhaka.
IT Brief Australia inoshuma 94% yeHumanEval kurongeka kwemamodheru-source-code kudzoreredza, zvichienzaniswa nenhamba dze48% yeAnthropic's Fable uye 45% yeOpenAI's GPT-5.5. Chirevo ichi chinotora RevEng.AI mukuru mukuru James Patrick-Evans uye Enterprise Management Associates muongorori Chris Steffen. Hapana mhedzisiro yebvunzo yeruzhinji, nzira yekuongorora, mienzaniso yevatengi, mitengo, kana mazwi ekuwana anopihwa mune kwakabva.
Kwakabva mashoko: itbrief.com.au ↗
Nei zvichikosha
Binary ongororo yakakosha pakuongorora wechitatu-bato, yakavharwa-sosi, yakagarwa nhaka, kana inogona kukanganisa software kana yekutanga kodhi kodhi isipo. Kana hunyanzvi hwakataurwa huchiita zvakavimbika kunze kwemabhenji akachekwa, vanogona kubatsira zvikwata zvekuchengetedza kuongorora njodzi dzekugovera-ketani, malware, uye software yekugadzira nekukurumidza. Nekudaro, chinyorwa chinopa RevEng.AI nhamba dzekuita uye kuenzanisa semhedzisiro yakashumwa nekambani, pasina kuyedza yakazvimirira kana nzira yakatsanangurwa.
Kuvhurwa kwacho kunogadzirisa dambudziko rakati rekuchengetedza: kuongorora software mushure mekubatanidza kana repository, zviratidzo, zvinotsamira, kana yepakutanga chinangwa chekuvandudza chingave chisipo. General-chinangwa mitauro modhi uye yakajairwa decompilers inogona kuburitsa isina kukwana kana yakaoma-kududzira mibairo, maererano neshumo. Sitimu yakasarudzika inogona kuderedza nguva inopedzwa nevanoongorora vachiongorora mabhinari asina kujairika, kunze kwekunge yabuda ikabatwa serubatsiro rwekuferefeta kwete humbowo hwekuchengetedza.
Kukosha kunoshanda kunoenderana nekuvimbika mumamiriro ezvinhu akaoma. Vanyori veMalware vanogona kushandisa kurongedza, obfuscation, isina kujairika kuunganidza marongero, uye runtime maitiro ayo static mabhinari ongororo anogona kusatora. Sosi yacho haipe humbowo hwakazvimirira nezvenyaya idzi, uye hairatidze kuti kodhi yakadzoserwa yakaenzana mukuchengetedzeka-hunoenderana nemaitiro. Iyo yakashumwa kuenzanisa saka inoratidza chikumbiro chekambani, kwete yakasimbiswa yekushanda mukana.
Interactive Mechanism: Iyo Inonyatsoshanda
Ongorora ari pasi tekinoroji kuseri kwekusimudzira uku uchipindirana.
Which component of an AI application is the machine-learning model itself?
Zvekutarisa zvinotevera
Mibvunzo yakakosha ndeyekuti Mega Bite inowanikwa kune vatengi, kuti inobatana sei neiripo kuchengetedza workflows, uye kana ongororo dzakazvimiririra dzichiburitsa mhedzisiro yakataurwa. Zvikwata zvekuchengetedza zvinofanirwawo kutarisa humbowo nezve manyepo, maitiro akapotsa akashata, kubata kweakabviswa kana kubvongodzwa mabhinari, kubata data, uye miganho yekodhi yakadzoserwa.
Sosi yacho haitaure kana Mega Bite inowanzo kuwanikwa, inongogumira kune bhizinesi kana vatengi vekudzivirira, inopihwa kuburikidza neAPI, kana inongotenderwa kune vakasarudzwa vashandisi. Izvo zvakare hazvipi mitengo, sarudzo dzekutumira, mazwi ekuchengetedza data, kana ruzivo rwekuti vatengi mabhinari anosiya nharaunda yesangano.
Kumwe kushuma kunofanirwa kuongorora yakazvimiririra bvunzo pamabhinari akabviswa, akagadziridzwa, akabatikana, uye ane hutsinye; mazinga ekuenzanisa kusina kururama; kunaka kwetsananguro dzakadzokororwa; uye kuti vaongorori vanosimbisa sei mhedziso dzemienzaniso. Humbowo kubva mukutumirwa mukupindura kwechiitiko kana software-simbiso workflows yaizove inodzidzisa kupfuura mabhenji emhedzisiro chete.
Zvikwata zvekuchengetedza zvinofunga nezvechigadzirwa zvinofanirwa kutsvaga magwaro akajeka pamusoro pezvivakwa zvinotsigirwa, mhando dzefaira, zvidzoreso zvekuvanzika, kutarisika, zvinodikanwa zvekuongorora kwevanhu, uye maratidziro ehurongwa hwekusagadzikana. Hapana yeaya ekuwana kana ruzivo rwekushanda inosimbiswa nekwakabva.