{"id":13797,"date":"2025-05-02T13:53:54","date_gmt":"2025-05-02T09:53:54","guid":{"rendered":"https:\/\/bte.iliauni.edu.ge\/?p=13797"},"modified":"2025-05-05T11:04:28","modified_gmt":"2025-05-05T07:04:28","slug":"importance-and-relevance-of-embedding-models-their-construction-and-the-methodology-for-selecting-the-appropriate-architecture","status":"publish","type":"post","link":"https:\/\/bte.iliauni.edu.ge\/en\/importance-and-relevance-of-embedding-models-their-construction-and-the-methodology-for-selecting-the-appropriate-architecture\/","title":{"rendered":"Importance and relevance of embedding models, their construction, and the methodology for selecting the appropriate architecture"},"content":{"rendered":"<p>On May 3, Ilia State University will host an event as part of Tbilisi AI Meetups:<\/p>\n<p>&#8220;Importance and relevance of embedding models, their construction, and the methodology for selecting the appropriate architecture.&#8221;<\/p>\n<p>In the age of powerful large language models, embeddings remain a critical component in retrieval, recommendation, and understanding systems. This session explores why embeddings still matter, how they\u2019re built, and how to choose the right architecture for your domain and application.<\/p>\n<p><strong>Agenda Highlights:<\/strong><\/p>\n<ul>\n<li>Context length limitations and the rise of RAG;<\/li>\n<li>Why LLMs aren\u2019t ideal for embedding generation;<\/li>\n<li>Causal attention vs contrastive learning;<\/li>\n<li>Sparse vs dense retrieval (TF-IDF, SPLADE, hybrid approaches);<\/li>\n<li>Bi-Encoders vs Cross-Encoders vs Late Interaction;<\/li>\n<li>Latent interaction models and HNSW search;<\/li>\n<li>Contrastive learning, triple loss, Siamese networks;<\/li>\n<li>Speed vs accuracy trade-offs;<\/li>\n<li>Domain-specific embedding tuning (e.g., code vs text);<\/li>\n<li>Evaluation with BEIR and MTEB benchmarks.<\/li>\n<\/ul>\n<div>\n<p><strong>Speaker:<\/strong>\u00a0Sandro Barnabishvili<\/p>\n<p><strong>Date &amp; Time:<\/strong>\u00a0Saturday, May 3, 13:00<\/p>\n<p>Register Here:\u00a0<a href=\"https:\/\/lu.ma\/n16g786u\">https:\/\/lu.ma\/n16g786u<\/a><\/p>\n<p><strong>Location:<\/strong>\u00a0Ilia State University, T Building, Auditorium #102<\/p>\n<p>2025<\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>On May 3, Ilia State University will host an event as part of Tbilisi AI Meetups: &#8220;Importance and relevance of embedding models, their construction, and the methodology for selecting the appropriate architecture.&#8221; In the age of powerful large language models, embeddings remain a critical component in retrieval, recommendation, and understanding systems. This session explores why&#8230;<\/p>\n","protected":false},"author":7,"featured_media":13805,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[172],"tags":[],"acf":[],"_links":{"self":[{"href":"https:\/\/bte.iliauni.edu.ge\/en\/wp-json\/wp\/v2\/posts\/13797"}],"collection":[{"href":"https:\/\/bte.iliauni.edu.ge\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/bte.iliauni.edu.ge\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/bte.iliauni.edu.ge\/en\/wp-json\/wp\/v2\/users\/7"}],"replies":[{"embeddable":true,"href":"https:\/\/bte.iliauni.edu.ge\/en\/wp-json\/wp\/v2\/comments?post=13797"}],"version-history":[{"count":0,"href":"https:\/\/bte.iliauni.edu.ge\/en\/wp-json\/wp\/v2\/posts\/13797\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/bte.iliauni.edu.ge\/en\/wp-json\/wp\/v2\/media\/13805"}],"wp:attachment":[{"href":"https:\/\/bte.iliauni.edu.ge\/en\/wp-json\/wp\/v2\/media?parent=13797"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/bte.iliauni.edu.ge\/en\/wp-json\/wp\/v2\/categories?post=13797"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/bte.iliauni.edu.ge\/en\/wp-json\/wp\/v2\/tags?post=13797"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}