Kwenzekeni
I-Herald Economy ibika ukuthi i-LG AI Research yethule amasistimu e-EXAONE okukhiqiza, isayensi kanye nezezimali ekhonsathini yayo ye-LG AI Talk Concert 2026. Umbiko uthi amasistimu ahlanganisa ukuqagela inqubo yefekthri, ukuhlola okubonakalayo, ukutholwa kwezinto ezibonakalayo, ukuhlola okuzimele kanye nokuhlaziywa kwezimali kwama-ejenti amaningi. Iphinde ichaze ukusebenzisana nokuthunyelwa, okuhlanganisa ukuhlanganiswa kwe-EXAONE 4.5 ohlelweni lokubuyekezwa kwezidakamizwa lwaseNingizimu Korea. Ukufinyelela komphakathi, amanani kanye nemiphumela yokusebenza okuzimele akunikezwanga.
I-Herald Economy ibika ukuthi Ucwaningo lwe-LG AI lubonise izinhlelo ze-AI zochwepheshe emcimbini obuseSeoul. Ekukhiqizeni, yethule i-EXAONE Tabular yokubikezela inqubo nekhwalithi kanye ne-EXAONE Omni-Inspect ukuze ihlolwe okubukwayo ngaphansi kokushintsha kwezimo zesithombe. Umbiko uthi bobabili bangasebenza kusuka enanini elincane ledatha esendaweni, kodwa abanikezeli izilinganiso ezizimele zokuqinisekisa lezo zimangalo.
Umbiko uthi i-LG ithuthukisa i-ejenti yokuhlola umbono ye-AI ehloselwe ukuphatha ukusampula kwedatha, ukulebula nokuqeqeshwa kwamamodeli, ngenhloso yesikhathi eside yezimboni ezizimele. Kwezesayensi, yethule i-EXAONE Discovery yokwakheka kwezinto kanye nokubikezela komphumela futhi yachaza umsebenzi ohlelweni lokuhlola oluzimele oluxhuma ukwakheka kwe-AI nokuhlolwa kwelebhu ebonakalayo.
I-Herald Economy iphinde ibike ukuthi i-LG yenza izinto zoketshezi olupholile nge-GS Caltex kanye namakhandidethi ezidakamizwa ze-oral peptide nge-D&D Pharmatech. Ithi i-EXAONE 4.5 ihlanganiswe ohlelweni lwe-AI loMnyango Wezokudla Nezidakamizwa WaseNingizimu Korea ukuze kubuyekezwe izidakamizwa ezintsha, nakuba noma yimuphi umthelela emigqeni yesikhathi yokugunyazwa uhlala ulindelekile kunomphumela obikiwe.
Kwezezimali, indatshana ithi i-EXAONE Business Intelligence isebenzisa ama-ejenti asebenzisanayo amaningi ekuhlaziyeni idatha, ukucabanga, ukubikezela kanye nencazelo. Kubikwa ukuthi yethulwe ekuqaleni kwalo nyaka futhi inwetshwa neLondon Stock Exchange Group kanye ne-Koscom, ngezinhlelo zezifunda ezengeziwe namakilasi ezimpahla. Ayikho imigomo yokufinyelela noma izintengo ezishiwo.
Imininingwane yomthombo: mbiz.heraldcorp.com ↗
Kungani kubalulekile
Lo mbiko uchaza ukusuka osizweni lwe-AI lwenhloso evamile ukuya kumasistimu ahloselwe ukuhumusha izimo ezishintshayo zezimboni, ukwenza izinqumo zokusebenza kanye nokuthatha izinyathelo. Uma amakhono abikiwe esebenza, angathinta ukulawulwa kwekhwalithi yefekthri, ucwaningo lwaselabhorethri, ukubuyekezwa kwezidakamizwa nokuhlaziywa kwezimali. Ukubaluleka kusalokhu kuzokwenzeka: I-Herald Economy ibika izimangalo nezinhlelo ze-LG, kodwa ayinikezi ukuhlola okuzimele, ukunemba okulinganiselwe, irekhodi lokuphepha noma ubufakazi bokuthi lezi zinhlelo zitholakala kabanzi.
Amasistimu abikiwe aqondise ukuhamba komsebenzi okulandelanayo lapho amaphutha engadala khona izinkinga zokukhiqiza, izibikezelo ezidukisayo, izenzo ezingaphephile zaselabhorethri noma izinkinga zokulawula. Ngakho-ke inani labo lincike ekuthembekeni, ekuhloleni nasekulawuleni abantu okuphumelelayo, okungekho okusungulwe yi-athikili.
Ukugcizelela kombiko kokuhlukile kanye nezimo ezishintshayo kubalulekile: amasistimu wezimboni kufanele asingathe izimo ezihlukile kudatha yokuqeqeshwa. Kodwa-ke, i-athikili ayizivezi izindlela zokuqinisekisa, izilinganiso zokwehluleka, izehlakalo zokuphepha noma ukuqhathanisa namathuluzi akhona.
I-athikili iphinda ibonise ukuthi amasistimu anamazinga ahlukene okuvuthwa. Ezinye zichazwa njengezihlanganisiwe noma eziqalisiwe, kuyilapho izimboni ezizimele, ukuhlola okuzimele kanye namakhono amarobhothi achazwa njengezinhloso zokuthuthukisa. Akufanele zithathwe njengezingahanjiswa ngokulinganayo.
I-Interactive Mechanism: Indlela Esebenza Ngayo Ngempela
Hlola ubuchwepheshe obuyisisekelo ngemuva kwalokhu kuthuthukiswa ngokuhlanganyela.
Which component of an AI application is the machine-learning model itself?
Ongakubuka ngokulandelayo
Buka imibhalo yezobuchwepheshe eyisisekelo, ukuhlaziya okuzimele kanye nobufakazi bokusetshenziswa kwangempela. Okungaziwa okubalulekile kufaka phakathi ukuthi yiziphi izinhlangano ezingasebenzisa isistimu ngayinye, noma ngabe yimuphi umkhiqizo uvame ukutholakala, amanani, idatha kanye nezidingo zengqalasizinda, ukugada komuntu, kanye nezilawuli zokuphepha ezinqumweni ezizimele. Ukwandiswa okuhleliwe kwe-LG kokufakwa kwezimali kanye nelabhorethri ezimele nomsebenzi wefekthri kufanele kuthathwe njengentuthuko yesikhathi esizayo kuze kube yilapho inkampani noma abasebenzisi bebhala imiphumela.
Ayikho inqubo yokubhalisa yomphakathi, isitatimende sokutholakala okujwayelekile noma amanani abhalwe phansi kumasistimu achaziwe. Ubambiswano olubikiwe nokuhlanganiswa kukahulumeni akubonisi ukuthi inkampani ngayinye, abacwaningi noma umphakathi ungafinyelela kukho.
Ukubika okwengeziwe kufanele kuveze ukuthi amasistimu asebenza kudatha yekhasimende, yiziphi izivikelo ezibusa izenzo ezizimele, nokuthi abasebenzisi bangakwazi yini ukuhlola noma babhale ngaphezulu izinqumo zemodeli.
Ukuhlola okuzimele kungaba usizo ikakhulukazi ekuhloleni okushintsha isithombe, ukubikezela kochungechunge lwesikhathi, izinto zokwakha nokutholwa kwezidakamizwa, nokuhlaziywa kwezimali kwama-agent amaningi. Umbiko awunikezi imiphumela enjalo.
Isihloko sithi i-LG ihlela ukunweba i-EXAONE Business Intelligence ibe yizifunda eziningi namakilasi ezimpahla, kodwa ayinikezi uhlelo lwezikhathi noma ubufakazi bokuthi lokho kwandiswa kwenzeke.