Aqoon-bixinta Qadiimiga ah: Modelka AI-ga ee Greek-ga ee Adduunka oo Ku Biiray Akhrinta Qoraallada Qadiimiga ah
Warshadda Sayniska Austria ayaa maanta soo bandhigaysa modelka AI-ga ee adduunka ugu horreeya ee loo sameeyay luuqadda Greek-ga, kaas oo loogu talagalay inuu caawiyo macallimiinta inay dib u soo ceshadaan qoraallada qadiimiga ah ee xad-dhaafka ah. Modelkan, oo loo magac baxay Apollo, wuxuu ku baranayaa 600 milyan oo eray oo Greek ah, waxaana loo abuuray inuu noqdo mid bilaash ah oo macallimiinta isticmaali karaan.
Sidee Modelkan U Shaqeeyaa?
Modelkan Apollo wuxuu ku baranayaa 600 milyan oo eray oo Greek ah, waxaana loo abuuray inuu noqdo mid bilaash ah oo macallimiinta isticmaali karaan. Wuxuu si toos ah u shaqeeyaa marka uu arko qoraal ka mid ah Homer, wuxuuna ku xoojiyaa luuqadda Homer, halka uu marka kale arko qoraal ka mid ah Doric, wuxuuna ku xoojiyaa luuqaddaas.
Dhanka kale, modelkan wuxuu u shaqeeyaa inuu soo jeediyo doorashooyin erayo ah, si macallimadu u dooran karaan kuwa saxda ah. Tani waxay ka dhigan tahay in macallimadu uusan u baahnayn inuu si gaar ah u ogaado erayada, balse uu u baahan yahay inuu dooran karo kuwa saxda ah.
Faa'iidooyinka Macallimiinta
Macallimiinta ayaa sheegaya in modelkan uu ka dhigi karo inay si fudud u ogaadaan qoraallada qadiimiga ah, iyagoo uusan u baahnayn inay si gaar ah u ogaadaan erayada. Tani waxay ka dhigan tahay in macallimiinta ay si fudud u ogaadaan qoraallada qadiimiga ah, iyagoo uusan u baahnayn inay si gaar ah u ogaadaan erayada.
Dhanka kale, modelkan wuxuu u shaqeeyaa inuu soo jeediyo doorashooyin erayo ah, si macallimadu u dooran karaan kuwa saxda ah. Tani waxay ka dhigan tahay in macallimiinta ay si fudud u ogaadaan qoraallada qadiimiga ah, iyagoo uusan u baahnayn inay si gaar ah u ogaadaan erayada.
Xisaabta Macallimiinta
Macallimiinta ayaa sheegaya in modelkan uu ka dhigi karo inay si fudud u ogaadaan qoraallada qadiimiga ah, iyagoo uusan u baahnayn inay si gaar ah u ogaadaan erayada. Tani waxay ka dhigan tahay in macallimiinta ay si fudud u ogaadaan qoraallada qadiimiga ah, iyagoo uusan u baahnayn inay si gaar ah u ogaadaan erayada.
Dhanka kale, modelkan wuxuu u shaqeeyaa inuu soo jeediyo doorashooyin erayo ah, si macallimadu u dooran karaan kuwa saxda ah. Tani waxay ka dhigan tahay in macallimiinta ay si fudud u ogaadaan qoraallada qadiimiga ah, iyagoo uusan u baahnayn inay si gaar ah u ogaadaan erayada.
On Wednesday, the Austrian Academy of Science will launch Apollo, the world's first advanced large language model dedicated to Ancient Greek, developed in partnership with French AI lab Mistral and technology firm Sail Reply.
Training on Hundreds of Millions of Words
Apollo is trained on roughly 600 million historical Greek words drawn from manuscripts, papyri, and inscriptions, giving it a deep statistical understanding of the language's nuances across different eras and dialects.
Until now, restoring a tattered piece of papyrus has required a skilled academic to first identify the word divisions, then accurately date the document, weigh the appropriate socio-political contexts, and consult reference materials to help choose suitable words to fill in the gaps.
Designed as a Scholarly Assistant, Not a Replacement
Apollo is built to propose a selection of word options for a scholar to select between, ensuring that human expertise remains central to the interpretation of historical material rather than being superseded by automated output.
The model will be freely available to academics through a chatbot interface, allowing researchers to more rapidly identify papyrus fragments relevant to their specific sub-disciplines and promising new avenues of research.
Potential to Transform Other Ancient Languages
If Apollo is a success, the same technique could be readily applied to other ancient languages—Latin or Egyptian, say—or any other academic discipline that would benefit from the distillation and indexing of a large corpus of material.
Apollo is unlikely to change the broad-strokes understanding of the ancient world; many papyri are yet to be restored precisely because they are mundane—personal letters, marital contracts, civil service papers.
Ilaha iyo xuquuqda sawirka
Sawir: Wired Xigasho
