{"id":224256,"date":"2025-11-02T19:19:34","date_gmt":"2025-11-02T20:19:34","guid":{"rendered":"https:\/\/express24.ir\/d\/product\/2112-07198\/"},"modified":"2025-11-02T19:19:34","modified_gmt":"2025-11-02T20:19:34","slug":"2112-07198","status":"publish","type":"product","link":"https:\/\/express24.ir\/d\/product\/2112-07198\/","title":{"rendered":"\u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0628\u0647 \u0647\u0645\u0631\u0627\u0647 PDF \u0641\u0627\u0631\u0633\u06cc + \u067e\u0627\u062f\u06a9\u0633\u062a \u0635\u0648\u062a\u06cc \u0641\u0627\u0631\u0633\u06cc + \u0648\u06cc\u062f\u06cc\u0648 \u0622\u0645\u0648\u0632\u0634\u06cc \u0641\u0627\u0631\u0633\u06cc"},"content":{"rendered":"<div style=\"background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); padding: 30px; border-radius: 15px; color: white; margin-bottom: 30px;\">\n<h2 style=\"color: white; text-align: center; margin-bottom: 20px;\">\ud83d\udcda \u0645\u0642\u0627\u0644\u0647 \u0639\u0644\u0645\u06cc<\/h2>\n<\/p><\/div>\n<table class=\"paper-info-table\" border=\"1\" cellpadding=\"10\" cellspacing=\"0\" style=\"width: 100%; margin-bottom: 20px; border-collapse: collapse; direction: rtl;\">\n<tr>\n<th style=\"background-color: #1F3A5E; color: white; text-align: right; width: 35%;\">\u0639\u0646\u0648\u0627\u0646 \u0641\u0627\u0631\u0633\u06cc \u0645\u0642\u0627\u0644\u0647<\/th>\n<td style=\"text-align: right; background-color: #f8f9fa; font-weight: bold;\">\u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647<\/td>\n<\/tr>\n<tr>\n<th style=\"background-color: #1F3A5E; color: white; text-align: right;\">\u0646\u0648\u06cc\u0633\u0646\u062f\u06af\u0627\u0646<\/th>\n<td style=\"text-align: right; background-color: #ffffff;\">Runxin Xu, Fuli Luo, Chengyu Wang, Baobao Chang, Jun Huang, Songfang Huang, Fei Huang<\/td>\n<\/tr>\n<tr>\n<th style=\"background-color: #1F3A5E; color: white; text-align: right;\">\u062f\u0633\u062a\u0647\u200c\u0628\u0646\u062f\u06cc \u0639\u0644\u0645\u06cc<\/th>\n<td style=\"text-align: right; background-color: #f8f9fa;\">Computation and Language,Artificial Intelligence<\/td>\n<\/tr>\n<\/table>\n<div style=\"border: 2px dashed #4CAF50; border-radius: 16px; padding: 20px; background: #f9fff9; font-family: 'IRANSans', sans-serif;\">\r\n  <h2 style=\"color: #2E7D32; margin-top: 0;\">\ud83d\udcd8 \u0645\u062d\u062a\u0648\u0627\u06cc \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647 \u0622\u0645\u0648\u0632\u0634\u06cc<\/h2>\r\n  <ul style=\"list-style-type: \u2705; padding-left: 20px; font-size: 16px; line-height: 1.8;\">\r\n    <li>\u0634\u0627\u0645\u0644 <strong>\u0641\u0627\u06cc\u0644 \u0627\u0635\u0644\u06cc \u0645\u0642\u0627\u0644\u0647 (PDF \u0627\u0646\u06af\u0644\u06cc\u0633\u06cc)<\/strong><\/li>\r\n    <li>\u0628\u0647 \u0647\u0645\u0631\u0627\u0647 <strong>\u0641\u0627\u06cc\u0644 PDF \u062a\u0648\u0636\u06cc\u062d \u0641\u0627\u0631\u0633\u06cc<\/strong> \u0628\u0627 \u0628\u06cc\u0627\u0646 \u0633\u0627\u062f\u0647 \u0648 \u0631\u0648\u0627\u0646<\/li>\r\n    <li>\u062f\u0627\u0631\u0627\u06cc <strong>\u067e\u0627\u062f\u06a9\u0633\u062a \u0635\u0648\u062a\u06cc \u0641\u0627\u0631\u0633\u06cc<\/strong> \u062a\u0648\u0636\u06cc\u062d \u06a9\u0627\u0645\u0644 \u0645\u0642\u0627\u0644\u0647<\/li>\r\n    <li>\u0628\u0647 \u0647\u0645\u0631\u0627\u0647 <strong>\u0648\u06cc\u062f\u06cc\u0648 \u0622\u0645\u0648\u0632\u0634\u06cc \u0641\u0627\u0631\u0633\u06cc<\/strong> \u0628\u0631\u0627\u06cc \u062f\u0631\u06a9 \u0639\u0645\u06cc\u0642\u200c\u062a\u0631 \u0645\u0641\u0627\u0647\u06cc\u0645 \u0645\u0642\u0627\u0644\u0647<\/li>\r\n  <\/ul>\r\n  <p style=\"color: #388E3C; font-weight: bold; font-size: 18px; margin-top: 20px;\">\r\n    \ud83c\udfaf \u0647\u0645\u0647\u200c\u06cc \u0641\u0627\u06cc\u0644\u200c\u0647\u0627 \u0628\u0627 \u0647\u062f\u0641 \u062f\u0631\u06a9 \u0622\u0633\u0627\u0646 \u0648 \u0633\u0631\u06cc\u0639 \u0645\u0641\u0627\u0647\u06cc\u0645 \u0639\u0644\u0645\u06cc \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647 \u062a\u0647\u06cc\u0647 \u0634\u062f\u0647\u200c\u0627\u0646\u062f.\r\n  <\/p>\r\n  <p style=\"font-size: 16px; line-height: 1.8;\">\r\n    \u0686\u0646\u0627\u0646\u0686\u0647 \u062f\u0631 <strong>\u062f\u0627\u0646\u0644\u0648\u062f \u0641\u0627\u06cc\u0644\u200c\u0647\u0627<\/strong> \u0628\u0627 \u0645\u0634\u06a9\u0644\u06cc \u0645\u0648\u0627\u062c\u0647 \u0634\u062f\u06cc\u062f\u060c \u0644\u0637\u0641\u0627\u064b \u0627\u0632 \u0637\u0631\u06cc\u0642 \u0648\u0627\u062a\u0633\u200c\u0627\u067e \u0628\u0627 \u0634\u0645\u0627\u0631\u0647 \r\n    <strong>09395106248<\/strong> \u06cc\u0627 \u0627\u0632 \u0637\u0631\u06cc\u0642 \u0622\u06cc\u062f\u06cc \u062a\u0644\u06af\u0631\u0627\u0645 \r\n    <strong>@ma_limbs<\/strong> \u067e\u06cc\u0627\u0645 \u062f\u0647\u06cc\u062f \u062a\u0627 \u0644\u06cc\u0646\u06a9\u200c\u0647\u0627 \u0641\u0648\u0631\u0627\u064b \u0628\u0631\u0627\u06cc\u062a\u0627\u0646 \u0645\u062c\u062f\u062f\u0627\u064b \u0627\u0631\u0633\u0627\u0644 \u0634\u0648\u0646\u062f.\r\n  <\/p>\r\n<\/div>\r\n\n<article class=\"violin-complex-pieces\" style=\"font-family: 'Vazirmatn', sans-serif;color: #2E2E2E;line-height: 1.75;max-width: 800px;margin: 40px auto;padding: 30px;background: #FFFFFF;border-radius: 12px;box-shadow: 0 6px 20px rgba(0,0,0,0.05);\">\n<style>\n.violin-complex-pieces h1 { text-align: center; font-size: 2.8rem; color: #1F3A5E; margin-bottom: 24px; }\n.violin-complex-pieces h2 { position: relative; font-size: 1.9rem; color: #243A5E; margin-top: 36px; margin-bottom: 16px; padding-bottom: 6px; }\n.violin-complex-pieces h2::after { content: \"\"; position: absolute; left: 0; bottom: 0; width: 60px; height: 4px; background: linear-gradient(90deg,#4A90E2,#50E3C2); border-radius: 2px; }\n.section-box { background: #F7F9FC; border: 1px solid #E0E4E8; border-radius: 8px; padding: 24px; margin-top: 20px; transition: transform 0.2s, box-shadow 0.2s; }\n.section-box:hover { transform: translateY(-4px); box-shadow: 0 8px 24px rgba(42,77,102,0.1); }\n.section-box p, .section-box ul { margin-bottom: 12px; }\nul { margin-left: 24px; }\nul li { margin-bottom: 8px; }\n.highlight { background: #EAF4FF; color: #1F3A5E; padding: 3px 6px; border-radius: 4px; font-weight: bold; }\n<\/style>\n<h1>\u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647<\/h1>\n<div class=\"section-box\">\n<p>\u062f\u0631 \u0639\u0635\u0631 \u062d\u0627\u0636\u0631\u060c \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 (Pre-trained Language Models &#8211; PLMs) \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u06cc\u06a9\u06cc \u0627\u0632 \u0627\u0631\u06a9\u0627\u0646 \u0627\u0635\u0644\u06cc \u062f\u0631 \u062d\u0648\u0632\u0647\u200c\u06cc \u067e\u0631\u062f\u0627\u0632\u0634 \u0632\u0628\u0627\u0646 \u0637\u0628\u06cc\u0639\u06cc (Natural Language Processing &#8211; NLP) \u0634\u0646\u0627\u062e\u062a\u0647 \u0645\u06cc\u200c\u0634\u0648\u0646\u062f. \u0627\u06cc\u0646 \u0645\u062f\u0644\u200c\u0647\u0627 \u0628\u0627 \u0628\u0647\u0631\u0647\u200c\u06af\u06cc\u0631\u06cc \u0627\u0632 \u062d\u062c\u0645 \u0639\u0638\u06cc\u0645\u06cc \u0627\u0632 \u062f\u0627\u062f\u0647\u200c\u0647\u0627\u06cc \u0645\u062a\u0646\u06cc\u060c \u062a\u0648\u0627\u0646\u0627\u06cc\u06cc \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u0627\u0644\u06af\u0648\u0647\u0627 \u0648 \u062f\u0627\u0646\u0634 \u0632\u0628\u0627\u0646\u06cc \u067e\u06cc\u0686\u06cc\u062f\u0647 \u0631\u0627 \u0628\u0647 \u062f\u0633\u062a \u0645\u06cc\u200c\u0622\u0648\u0631\u0646\u062f \u0648 \u062f\u0631 \u0637\u06cc\u0641 \u06af\u0633\u062a\u0631\u062f\u0647\u200c\u0627\u06cc \u0627\u0632 \u0648\u0638\u0627\u06cc\u0641 NLP\u060c \u0627\u0632 \u062a\u0631\u062c\u0645\u0647 \u0645\u0627\u0634\u06cc\u0646\u06cc \u06af\u0631\u0641\u062a\u0647 \u062a\u0627 \u062a\u062d\u0644\u06cc\u0644 \u0627\u062d\u0633\u0627\u0633\u0627\u062a\u060c \u0628\u0647 \u06a9\u0627\u0631 \u06af\u0631\u0641\u062a\u0647 \u0645\u06cc\u200c\u0634\u0648\u0646\u062f. \u0628\u0627 \u0627\u06cc\u0646 \u062d\u0627\u0644\u060c \u0627\u06cc\u0646 \u0642\u062f\u0631\u062a \u0648 \u062a\u0648\u0627\u0646\u0627\u06cc\u06cc\u060c \u0628\u0647 \u0628\u0647\u0627\u06cc \u0633\u0646\u06af\u06cc\u0646\u06cc \u0627\u0632 \u0646\u0638\u0631 \u0645\u062d\u0627\u0633\u0628\u0627\u062a\u06cc \u0648 \u0645\u0635\u0631\u0641 \u0645\u0646\u0627\u0628\u0639 \u0633\u062e\u062a\u200c\u0627\u0641\u0632\u0627\u0631\u06cc \u0628\u0647 \u062f\u0633\u062a \u0645\u06cc\u200c\u0622\u06cc\u062f. \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc \u0628\u0632\u0631\u06af\u060c \u0628\u0627 \u0645\u06cc\u0644\u06cc\u0627\u0631\u062f\u0647\u0627 \u067e\u0627\u0631\u0627\u0645\u062a\u0631\u060c \u0646\u06cc\u0627\u0632\u0645\u0646\u062f \u0645\u0646\u0627\u0628\u0639 \u0645\u062d\u0627\u0633\u0628\u0627\u062a\u06cc \u0642\u0627\u0628\u0644 \u062a\u0648\u062c\u0647\u06cc \u0628\u0631\u0627\u06cc \u0622\u0645\u0648\u0632\u0634\u060c \u0627\u062c\u0631\u0627 \u0648 \u0627\u0633\u062a\u0642\u0631\u0627\u0631 \u0647\u0633\u062a\u0646\u062f. \u0627\u06cc\u0646 \u0627\u0645\u0631\u060c \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 \u0627\u06cc\u0646 \u0645\u062f\u0644\u200c\u0647\u0627 \u0631\u0627 \u062f\u0631 \u0645\u062d\u06cc\u0637\u200c\u0647\u0627\u06cc \u0628\u0627 \u0645\u062d\u062f\u0648\u062f\u06cc\u062a \u0645\u0646\u0627\u0628\u0639\u060c \u0645\u0627\u0646\u0646\u062f \u062f\u0633\u062a\u06af\u0627\u0647\u200c\u0647\u0627\u06cc \u062a\u0644\u0641\u0646 \u0647\u0645\u0631\u0627\u0647 \u06cc\u0627 \u0633\u06cc\u0633\u062a\u0645\u200c\u0647\u0627\u06cc \u062a\u0639\u0628\u06cc\u0647\u200c\u0634\u062f\u0647\u060c \u0628\u0627 \u0686\u0627\u0644\u0634\u200c\u0647\u0627\u06cc \u062c\u062f\u06cc \u0645\u0648\u0627\u062c\u0647 \u0645\u06cc\u200c\u0633\u0627\u0632\u062f.<\/p>\n<\/p><\/div>\n<h2>\u0627\u0647\u0645\u06cc\u062a \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc<\/h2>\n<div class=\"section-box\">\n<p>\u0628\u0647 \u0647\u0645\u06cc\u0646 \u062f\u0644\u06cc\u0644\u060c \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc\u060c \u0628\u0647 \u06cc\u06a9\u06cc \u0627\u0632 \u0632\u0645\u06cc\u0646\u0647\u200c\u0647\u0627\u06cc \u062a\u062d\u0642\u06cc\u0642\u0627\u062a\u06cc \u0641\u0639\u0627\u0644 \u0648 \u062d\u06cc\u0627\u062a\u06cc \u062f\u0631 NLP \u062a\u0628\u062f\u06cc\u0644 \u0634\u062f\u0647 \u0627\u0633\u062a. \u0647\u062f\u0641 \u0627\u0632 \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u060c \u06a9\u0627\u0647\u0634 \u062d\u062c\u0645 \u0648 \u067e\u06cc\u0686\u06cc\u062f\u06af\u06cc \u0645\u062f\u0644 \u0628\u062f\u0648\u0646 \u0627\u0641\u062a \u0642\u0627\u0628\u0644 \u062a\u0648\u062c\u0647 \u062f\u0631 \u0639\u0645\u0644\u06a9\u0631\u062f \u0622\u0646 \u0627\u0633\u062a. \u062a\u06a9\u0646\u06cc\u06a9\u200c\u0647\u0627\u06cc \u0645\u062e\u062a\u0644\u0641\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u200c\u0647\u0627 \u0648\u062c\u0648\u062f \u062f\u0627\u0631\u062f\u060c \u0627\u0632 \u062c\u0645\u0644\u0647:<\/p>\n<ul>\n<li><span class=\"highlight\">\u0647\u0631\u0633 (Pruning)<\/span>: \u062d\u0630\u0641 \u067e\u0627\u0631\u0627\u0645\u062a\u0631\u0647\u0627\u06cc \u063a\u06cc\u0631\u0636\u0631\u0648\u0631\u06cc \u0645\u062f\u0644.<\/li>\n<li><span class=\"highlight\">\u06a9\u0648\u0627\u0646\u062a\u06cc\u0632\u0627\u0633\u06cc\u0648\u0646 (Quantization)<\/span>: \u06a9\u0627\u0647\u0634 \u062f\u0642\u062a \u0646\u0645\u0627\u06cc\u0634 \u067e\u0627\u0631\u0627\u0645\u062a\u0631\u0647\u0627\u06cc \u0645\u062f\u0644.<\/li>\n<li><span class=\"highlight\">\u062a\u062e\u0644\u06cc\u0635 \u062f\u0627\u0646\u0634 (Knowledge Distillation)<\/span>: \u0622\u0645\u0648\u0632\u0634 \u06cc\u06a9 \u0645\u062f\u0644 \u06a9\u0648\u0686\u06a9\u062a\u0631 \u0628\u0631\u0627\u06cc \u062a\u0642\u0644\u06cc\u062f \u0631\u0641\u062a\u0627\u0631 \u06cc\u06a9 \u0645\u062f\u0644 \u0628\u0632\u0631\u06af\u062a\u0631.<\/li>\n<\/ul>\n<p>\u0647\u0631 \u06cc\u06a9 \u0627\u0632 \u0627\u06cc\u0646 \u0631\u0648\u0634\u200c\u0647\u0627\u060c \u0645\u0632\u0627\u06cc\u0627 \u0648 \u0645\u0639\u0627\u06cc\u0628 \u062e\u0627\u0635 \u062e\u0648\u062f \u0631\u0627 \u062f\u0627\u0631\u0646\u062f \u0648 \u0628\u0633\u062a\u0647 \u0628\u0647 \u0634\u0631\u0627\u06cc\u0637 \u0648 \u0646\u06cc\u0627\u0632\u0647\u0627\u06cc \u062e\u0627\u0635\u060c \u0645\u06cc\u200c\u062a\u0648\u0627\u0646 \u0627\u0632 \u0622\u0646\u200c\u0647\u0627 \u0628\u0647 \u0635\u0648\u0631\u062a \u062c\u062f\u0627\u06af\u0627\u0646\u0647 \u06cc\u0627 \u062a\u0631\u06a9\u06cc\u0628\u06cc \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u06a9\u0631\u062f.<\/p>\n<\/p><\/div>\n<h2>\u0645\u0639\u0631\u0641\u06cc \u0645\u0642\u0627\u0644\u0647<\/h2>\n<div class=\"section-box\">\n<p>\u0645\u0642\u0627\u0644\u0647 \u062d\u0627\u0636\u0631\u060c \u0628\u0627 \u0639\u0646\u0648\u0627\u0646 <span class=\"highlight\">&#8220;\u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647&#8221;<\/span> (From Dense to Sparse: Contrastive Pruning for Better Pre-trained Language Model Compression)\u060c \u0628\u0647 \u0628\u0631\u0631\u0633\u06cc \u0648 \u0628\u0647\u0628\u0648\u062f \u0631\u0648\u0634\u200c\u0647\u0627\u06cc \u0647\u0631\u0633 \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0645\u06cc\u200c\u067e\u0631\u062f\u0627\u0632\u062f. \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647\u060c \u06cc\u06a9 \u0631\u0648\u06cc\u06a9\u0631\u062f \u062c\u062f\u06cc\u062f \u0628\u0647 \u0646\u0627\u0645 <span class=\"highlight\">&#8220;\u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc&#8221;<\/span> (Contrastive Pruning &#8211; CAP) \u0631\u0627 \u0645\u0639\u0631\u0641\u06cc \u0645\u06cc\u200c\u06a9\u0646\u062f \u06a9\u0647 \u0647\u062f\u0641 \u0622\u0646\u060c \u062d\u0641\u0638 \u062f\u0627\u0646\u0634 \u0647\u0645\u06af\u0627\u0646\u06cc (task-agnostic) \u0648 \u062f\u0627\u0646\u0634 \u062e\u0627\u0635 \u0648\u0638\u06cc\u0641\u0647 (task-specific) \u062f\u0631 \u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647 \u0627\u0633\u062a. \u0627\u06cc\u0646 \u0631\u0648\u06cc\u06a9\u0631\u062f\u060c \u0628\u0627 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u062a\u0642\u0627\u0628\u0644\u06cc (Contrastive Learning)\u060c \u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647 \u0631\u0627 \u0642\u0627\u062f\u0631 \u0645\u06cc\u200c\u0633\u0627\u0632\u062f \u062a\u0627 \u0627\u0632 \u0645\u062f\u0644 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0648 \u0645\u062f\u0644 \u062a\u0646\u0638\u06cc\u0645\u200c\u0634\u062f\u0647 (fine-tuned) \u0628\u0647 \u0637\u0648\u0631 \u0647\u0645\u0632\u0645\u0627\u0646 \u06cc\u0627\u062f \u0628\u06af\u06cc\u0631\u062f.<\/p>\n<\/p><\/div>\n<h2>\u0646\u0648\u06cc\u0633\u0646\u062f\u06af\u0627\u0646 \u0648 \u0632\u0645\u06cc\u0646\u0647 \u062a\u062d\u0642\u06cc\u0642<\/h2>\n<div class=\"section-box\">\n<p>\u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647 \u062a\u0648\u0633\u0637 <span class=\"highlight\">Runxin Xu, Fuli Luo, Chengyu Wang, Baobao Chang, Jun Huang, Songfang Huang, \u0648 Fei Huang<\/span> \u0646\u0648\u0634\u062a\u0647 \u0634\u062f\u0647 \u0627\u0633\u062a. \u0627\u06cc\u0646 \u0646\u0648\u06cc\u0633\u0646\u062f\u06af\u0627\u0646\u060c \u0645\u062a\u062e\u0635\u0635\u0627\u0646 \u062d\u0648\u0632\u0647\u200c\u0647\u0627\u06cc \u0645\u062d\u0627\u0633\u0628\u0627\u062a \u0648 \u0632\u0628\u0627\u0646 (Computation and Language) \u0648 \u0647\u0648\u0634 \u0645\u0635\u0646\u0648\u0639\u06cc (Artificial Intelligence) \u0647\u0633\u062a\u0646\u062f \u0648 \u062a\u062c\u0631\u0628\u0647 \u0642\u0627\u0628\u0644 \u062a\u0648\u062c\u0647\u06cc \u062f\u0631 \u0632\u0645\u06cc\u0646\u0647 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0648 \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644 \u062f\u0627\u0631\u0646\u062f. \u062a\u062d\u0642\u06cc\u0642\u0627\u062a \u0622\u0646\u200c\u0647\u0627 \u0628\u0631 \u0628\u0647\u0628\u0648\u062f \u06a9\u0627\u0631\u0627\u06cc\u06cc \u0648 \u0642\u0627\u0628\u0644\u06cc\u062a \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc \u0628\u0632\u0631\u06af \u0645\u062a\u0645\u0631\u06a9\u0632 \u0627\u0633\u062a.<\/p>\n<\/p><\/div>\n<h2>\u0686\u06a9\u06cc\u062f\u0647 \u0648 \u062e\u0644\u0627\u0635\u0647 \u0645\u062d\u062a\u0648\u0627<\/h2>\n<div class=\"section-box\">\n<p>\u0686\u06a9\u06cc\u062f\u0647 \u0645\u0642\u0627\u0644\u0647 \u0628\u0647 \u0627\u06cc\u0646 \u0635\u0648\u0631\u062a \u0627\u0633\u062a: \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 (PLM) \u062f\u0631 \u0648\u0638\u0627\u06cc\u0641 \u0645\u062e\u062a\u0644\u0641 \u067e\u0631\u062f\u0627\u0632\u0634 \u0632\u0628\u0627\u0646 \u0637\u0628\u06cc\u0639\u06cc (NLP) \u062a\u062d\u062a \u0627\u0644\u06af\u0648\u06cc \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u0632\u0634 \u0648 \u062a\u0646\u0638\u06cc\u0645 \u062f\u0642\u06cc\u0642\u060c \u0628\u0647 \u0645\u0648\u0641\u0642\u06cc\u062a\u200c\u0647\u0627\u06cc \u0628\u0632\u0631\u06af\u06cc \u062f\u0633\u062a \u06cc\u0627\u0641\u062a\u0647\u200c\u0627\u0646\u062f. \u0645\u062f\u0644\u200c\u0647\u0627\u06cc PLM \u0628\u0627 \u062f\u0627\u0634\u062a\u0646 \u062a\u0639\u062f\u0627\u062f \u0632\u06cc\u0627\u062f\u06cc \u067e\u0627\u0631\u0627\u0645\u062a\u0631\u060c \u0627\u0632 \u0646\u0638\u0631 \u0645\u062d\u0627\u0633\u0628\u0627\u062a\u06cc \u0648 \u0645\u0646\u0627\u0628\u0639\u060c \u067e\u0631\u0647\u0632\u06cc\u0646\u0647 \u0647\u0633\u062a\u0646\u062f. \u0627\u0632 \u0627\u06cc\u0646 \u0631\u0648\u060c \u0647\u0631\u0633 \u0645\u062f\u0644 \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u200c\u0647\u0627\u06cc PLM \u062f\u0631 \u0645\u0642\u06cc\u0627\u0633 \u0628\u0632\u0631\u06af \u0645\u0639\u0631\u0641\u06cc \u0634\u062f\u0647 \u0627\u0633\u062a. \u0628\u0627 \u0627\u06cc\u0646 \u062d\u0627\u0644\u060c \u0628\u06cc\u0634\u062a\u0631 \u0631\u0648\u06cc\u06a9\u0631\u062f\u0647\u0627\u06cc \u0642\u0628\u0644\u06cc \u0641\u0642\u0637 \u062f\u0627\u0646\u0634 \u062e\u0627\u0635 \u0648\u0638\u06cc\u0641\u0647 \u0631\u0627 \u0646\u0633\u0628\u062a \u0628\u0647 \u0648\u0638\u0627\u06cc\u0641 \u067e\u0627\u06cc\u06cc\u0646\u200c\u062f\u0633\u062a\u06cc \u062f\u0631 \u0646\u0638\u0631 \u0645\u06cc\u200c\u06af\u06cc\u0631\u0646\u062f\u060c \u0627\u0645\u0627 \u062f\u0627\u0646\u0634 \u0636\u0631\u0648\u0631\u06cc \u0645\u0633\u062a\u0642\u0644 \u0627\u0632 \u0648\u0638\u06cc\u0641\u0647 \u0631\u0627 \u062f\u0631 \u0637\u0648\u0644 \u0647\u0631\u0633 \u0646\u0627\u062f\u06cc\u062f\u0647 \u0645\u06cc\u200c\u06af\u06cc\u0631\u0646\u062f\u060c \u06a9\u0647 \u0645\u0645\u06a9\u0646 \u0627\u0633\u062a \u0628\u0627\u0639\u062b \u0645\u0634\u06a9\u0644 \u0641\u0631\u0627\u0645\u0648\u0634\u06cc \u0641\u0627\u062c\u0639\u0647\u200c\u0628\u0627\u0631 \u0634\u0648\u062f \u0648 \u0645\u0646\u062c\u0631 \u0628\u0647 \u062a\u0648\u0627\u0646\u0627\u06cc\u06cc \u062a\u0639\u0645\u06cc\u0645 \u0636\u0639\u06cc\u0641 \u0634\u0648\u062f. \u0628\u0631\u0627\u06cc \u062d\u0641\u0638 \u062f\u0627\u0646\u0634 \u0645\u0633\u062a\u0642\u0644 \u0627\u0632 \u0648\u0638\u06cc\u0641\u0647 \u0648 \u062f\u0627\u0646\u0634 \u062e\u0627\u0635 \u0648\u0638\u06cc\u0641\u0647 \u062f\u0631 \u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647\u060c \u0645\u0627 \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc (CAP) \u0631\u0627 \u062a\u062d\u062a \u0627\u0644\u06af\u0648\u06cc \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u0632\u0634 \u0648 \u062a\u0646\u0638\u06cc\u0645 \u062f\u0642\u06cc\u0642 \u067e\u06cc\u0634\u0646\u0647\u0627\u062f \u0645\u06cc\u200c\u06a9\u0646\u06cc\u0645. \u0627\u06cc\u0646 \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u06cc\u06a9 \u0686\u0627\u0631\u0686\u0648\u0628 \u06a9\u0644\u06cc \u0637\u0631\u0627\u062d\u06cc \u0634\u062f\u0647 \u0627\u0633\u062a\u060c \u06a9\u0647 \u0628\u0627 \u0647\u0631\u0633 \u0633\u0627\u062e\u062a\u0627\u0631\u06cc\u0627\u0641\u062a\u0647 \u0648 \u0628\u062f\u0648\u0646 \u0633\u0627\u062e\u062a\u0627\u0631 \u0633\u0627\u0632\u06af\u0627\u0631 \u0627\u0633\u062a. CAP \u06a9\u0647 \u062f\u0631 \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u062a\u0642\u0627\u0628\u0644\u06cc \u0645\u062a\u062d\u062f \u0634\u062f\u0647 \u0627\u0633\u062a\u060c \u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647 \u0631\u0627 \u0642\u0627\u062f\u0631 \u0645\u06cc\u200c\u0633\u0627\u0632\u062f \u062a\u0627 \u0627\u0632 \u0645\u062f\u0644 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0628\u0631\u0627\u06cc \u062f\u0627\u0646\u0634 \u0645\u0633\u062a\u0642\u0644 \u0627\u0632 \u0648\u0638\u06cc\u0641\u0647 \u0648 \u0645\u062f\u0644 \u062a\u0646\u0638\u06cc\u0645 \u062f\u0642\u06cc\u0642 \u0628\u0631\u0627\u06cc \u062f\u0627\u0646\u0634 \u062e\u0627\u0635 \u0648\u0638\u06cc\u0641\u0647 \u06cc\u0627\u062f \u0628\u06af\u06cc\u0631\u062f. \u0639\u0644\u0627\u0648\u0647 \u0628\u0631 \u0627\u06cc\u0646\u060c \u0628\u0631\u0627\u06cc \u062d\u0641\u0638 \u0628\u0647\u062a\u0631 \u0639\u0645\u0644\u06a9\u0631\u062f \u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647\u060c \u0627\u0633\u0646\u067e\u200c\u0634\u0627\u062a\u200c\u0647\u0627 (\u06cc\u0639\u0646\u06cc \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0645\u06cc\u0627\u0646\u06cc \u062f\u0631 \u0647\u0631 \u062a\u06a9\u0631\u0627\u0631 \u0647\u0631\u0633) \u0646\u06cc\u0632 \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0646\u0638\u0627\u0631\u062a\u200c\u0647\u0627\u06cc \u0645\u0648\u062b\u0631 \u0628\u0631\u0627\u06cc \u0647\u0631\u0633 \u0639\u0645\u0644 \u0645\u06cc\u200c\u06a9\u0646\u0646\u062f. \u0622\u0632\u0645\u0627\u06cc\u0634\u200c\u0647\u0627\u06cc \u06af\u0633\u062a\u0631\u062f\u0647 \u0645\u0627 \u0646\u0634\u0627\u0646 \u0645\u06cc\u200c\u062f\u0647\u062f \u06a9\u0647 \u0627\u062a\u062e\u0627\u0630 CAP \u0628\u0647 \u0637\u0648\u0631 \u0645\u062f\u0627\u0648\u0645 \u0645\u0646\u062c\u0631 \u0628\u0647 \u0628\u0647\u0628\u0648\u062f\u0647\u0627\u06cc \u0642\u0627\u0628\u0644 \u062a\u0648\u062c\u0647\u06cc \u0645\u06cc\u200c\u0634\u0648\u062f\u060c \u0628\u0647 \u062e\u0635\u0648\u0635 \u062f\u0631 \u0633\u0646\u0627\u0631\u06cc\u0648\u0647\u0627\u06cc \u067e\u0631\u0627\u06a9\u0646\u062f\u06af\u06cc \u0628\u0633\u06cc\u0627\u0631 \u0628\u0627\u0644\u0627. \u0628\u0627 \u062a\u0646\u0647\u0627 \u06f3\u066a \u067e\u0627\u0631\u0627\u0645\u062a\u0631\u0647\u0627\u06cc \u0645\u062f\u0644 \u0631\u0632\u0631\u0648 \u0634\u062f\u0647 (\u06cc\u0639\u0646\u06cc \u06f9\u06f7\u066a \u067e\u0631\u0627\u06a9\u0646\u062f\u06af\u06cc)\u060c CAP \u0628\u0627 \u0645\u0648\u0641\u0642\u06cc\u062a \u0628\u0647 \u06f9\u06f9.\u06f2\u066a \u0648 \u06f9\u06f6.\u06f3\u066a \u0627\u0632 \u0639\u0645\u0644\u06a9\u0631\u062f \u0627\u0635\u0644\u06cc BERT \u062f\u0631 \u0648\u0638\u0627\u06cc\u0641 QQP \u0648 MNLI \u062f\u0633\u062a \u0645\u06cc\u200c\u06cc\u0627\u0628\u062f. \u0639\u0644\u0627\u0648\u0647 \u0628\u0631 \u0627\u06cc\u0646\u060c \u0622\u0632\u0645\u0627\u06cc\u0634\u200c\u0647\u0627\u06cc \u06a9\u0627\u0648\u0634 \u0645\u0627 \u0646\u0634\u0627\u0646 \u0645\u06cc\u200c\u062f\u0647\u062f \u06a9\u0647 \u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647 \u062a\u0648\u0633\u0637 CAP \u062a\u0645\u0627\u06cc\u0644 \u062f\u0627\u0631\u062f \u0628\u0647 \u062a\u0648\u0627\u0646\u0627\u06cc\u06cc \u062a\u0639\u0645\u06cc\u0645 \u0628\u0647\u062a\u0631\u06cc \u062f\u0633\u062a \u06cc\u0627\u0628\u062f.<\/p>\n<p>\u0628\u0647 \u0637\u0648\u0631 \u062e\u0644\u0627\u0635\u0647\u060c \u0645\u0642\u0627\u0644\u0647 CAP \u0631\u0627 \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u06cc\u06a9 \u0631\u0648\u0634 \u06a9\u0627\u0631\u0622\u0645\u062f \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0645\u0639\u0631\u0641\u06cc \u0645\u06cc\u200c\u06a9\u0646\u062f. \u0627\u06cc\u0646 \u0631\u0648\u0634\u060c \u0628\u0627 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u062a\u0642\u0627\u0628\u0644\u06cc\u060c \u0642\u0627\u062f\u0631 \u0627\u0633\u062a \u062a\u0639\u0627\u062f\u0644\u06cc \u0628\u06cc\u0646 \u062d\u0641\u0638 \u062f\u0627\u0646\u0634 \u0639\u0645\u0648\u0645\u06cc \u0648 \u062f\u0627\u0646\u0634 \u062e\u0627\u0635 \u0648\u0638\u06cc\u0641\u0647 \u0628\u0631\u0642\u0631\u0627\u0631 \u06a9\u0646\u062f \u0648 \u062f\u0631 \u0646\u062a\u06cc\u062c\u0647\u060c \u0639\u0645\u0644\u06a9\u0631\u062f \u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647 \u0631\u0627 \u0628\u0647 \u0637\u0648\u0631 \u0642\u0627\u0628\u0644 \u062a\u0648\u062c\u0647\u06cc \u0628\u0647\u0628\u0648\u062f \u0628\u062e\u0634\u062f.<\/p>\n<\/p><\/div>\n<h2>\u0631\u0648\u0634\u200c\u0634\u0646\u0627\u0633\u06cc \u062a\u062d\u0642\u06cc\u0642<\/h2>\n<div class=\"section-box\">\n<p>\u0631\u0648\u0634\u200c\u0634\u0646\u0627\u0633\u06cc \u062a\u062d\u0642\u06cc\u0642 \u062f\u0631 \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647\u060c \u0645\u0628\u062a\u0646\u06cc \u0628\u0631 \u0631\u0648\u06cc\u06a9\u0631\u062f \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u062a\u0642\u0627\u0628\u0644\u06cc \u0627\u0633\u062a. \u062f\u0631 \u0627\u06cc\u0646 \u0631\u0648\u06cc\u06a9\u0631\u062f\u060c \u062f\u0648 \u0646\u0648\u0639 \u062f\u0627\u0646\u0634 (\u0639\u0645\u0648\u0645\u06cc \u0648 \u062e\u0627\u0635) \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0646\u0645\u0648\u0646\u0647\u200c\u0647\u0627\u06cc \u0645\u062b\u0628\u062a \u0648 \u0645\u0646\u0641\u06cc \u062f\u0631 \u0646\u0638\u0631 \u06af\u0631\u0641\u062a\u0647 \u0645\u06cc\u200c\u0634\u0648\u0646\u062f. \u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647\u060c \u0622\u0645\u0648\u0632\u0634 \u062f\u0627\u062f\u0647 \u0645\u06cc\u200c\u0634\u0648\u062f \u062a\u0627 \u0646\u0645\u0648\u0646\u0647\u200c\u0647\u0627\u06cc \u0645\u062b\u0628\u062a \u0631\u0627 \u0627\u0632 \u0646\u0645\u0648\u0646\u0647\u200c\u0647\u0627\u06cc \u0645\u0646\u0641\u06cc \u062a\u0634\u062e\u06cc\u0635 \u062f\u0647\u062f. \u0627\u06cc\u0646 \u0641\u0631\u0622\u06cc\u0646\u062f\u060c \u0628\u0647 \u0645\u062f\u0644 \u06a9\u0645\u06a9 \u0645\u06cc\u200c\u06a9\u0646\u062f \u062a\u0627 \u062f\u0627\u0646\u0634 \u0645\u0647\u0645 \u0631\u0627 \u062d\u0641\u0638 \u06a9\u0631\u062f\u0647 \u0648 \u062f\u0627\u0646\u0634 \u063a\u06cc\u0631\u0636\u0631\u0648\u0631\u06cc \u0631\u0627 \u062d\u0630\u0641 \u06a9\u0646\u062f.<\/p>\n<p>\u0628\u0647 \u0637\u0648\u0631 \u062f\u0642\u06cc\u0642\u200c\u062a\u0631\u060c CAP \u0627\u0632 \u0633\u0647 \u062c\u0632\u0621 \u0627\u0635\u0644\u06cc \u062a\u0634\u06a9\u06cc\u0644 \u0634\u062f\u0647 \u0627\u0633\u062a:<\/p>\n<ul>\n<li><span class=\"highlight\">\u0645\u062f\u0644 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647<\/span>: \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0645\u0646\u0628\u0639 \u062f\u0627\u0646\u0634 \u0639\u0645\u0648\u0645\u06cc \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0645\u06cc\u200c\u0634\u0648\u062f.<\/li>\n<li><span class=\"highlight\">\u0645\u062f\u0644 \u062a\u0646\u0638\u06cc\u0645\u200c\u0634\u062f\u0647<\/span>: \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0645\u0646\u0628\u0639 \u062f\u0627\u0646\u0634 \u062e\u0627\u0635 \u0648\u0638\u06cc\u0641\u0647 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0645\u06cc\u200c\u0634\u0648\u062f.<\/li>\n<li><span class=\"highlight\">\u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647<\/span>: \u0645\u062f\u0644\u06cc \u06a9\u0647 \u0628\u0627\u06cc\u062f \u0641\u0634\u0631\u062f\u0647 \u0634\u0648\u062f \u0648 \u062f\u0631 \u0639\u06cc\u0646 \u062d\u0627\u0644\u060c \u0639\u0645\u0644\u06a9\u0631\u062f \u062e\u0648\u062f \u0631\u0627 \u062d\u0641\u0638 \u06a9\u0646\u062f.<\/li>\n<\/ul>\n<p>\u062f\u0631 \u0637\u0648\u0644 \u0641\u0631\u0622\u06cc\u0646\u062f \u0622\u0645\u0648\u0632\u0634\u060c \u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647 \u0628\u0627 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 \u06cc\u06a9 \u062a\u0627\u0628\u0639 \u0636\u0631\u0631 \u062a\u0642\u0627\u0628\u0644\u06cc\u060c \u0622\u0645\u0648\u0632\u0634 \u062f\u0627\u062f\u0647 \u0645\u06cc\u200c\u0634\u0648\u062f \u062a\u0627 \u062e\u0631\u0648\u062c\u06cc\u200c\u0647\u0627\u06cc \u062e\u0648\u062f \u0631\u0627 \u0628\u0627 \u062e\u0631\u0648\u062c\u06cc\u200c\u0647\u0627\u06cc \u0645\u062f\u0644 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0648 \u0645\u062f\u0644 \u062a\u0646\u0638\u06cc\u0645\u200c\u0634\u062f\u0647 \u0647\u0645\u0627\u0647\u0646\u06af \u06a9\u0646\u062f. \u0627\u06cc\u0646 \u0641\u0631\u0622\u06cc\u0646\u062f\u060c \u0628\u0647 \u0645\u062f\u0644 \u06a9\u0645\u06a9 \u0645\u06cc\u200c\u06a9\u0646\u062f \u062a\u0627 \u0647\u0645 \u062f\u0627\u0646\u0634 \u0639\u0645\u0648\u0645\u06cc \u0648 \u0647\u0645 \u062f\u0627\u0646\u0634 \u062e\u0627\u0635 \u0648\u0638\u06cc\u0641\u0647 \u0631\u0627 \u0628\u0647 \u0637\u0648\u0631 \u0647\u0645\u0632\u0645\u0627\u0646 \u06cc\u0627\u062f \u0628\u06af\u06cc\u0631\u062f.<\/p>\n<p>\u0639\u0644\u0627\u0648\u0647 \u0628\u0631 \u0627\u06cc\u0646\u060c \u0645\u0642\u0627\u0644\u0647 \u0627\u0632 <span class=\"highlight\">\u0627\u0633\u0646\u067e\u200c\u0634\u0627\u062a\u200c\u0647\u0627<\/span> (\u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0645\u06cc\u0627\u0646\u06cc \u062f\u0631 \u0637\u0648\u0644 \u0641\u0631\u0622\u06cc\u0646\u062f \u0647\u0631\u0633) \u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0646\u0648\u0639\u06cc \u0646\u0638\u0627\u0631\u062a \u0627\u0636\u0627\u0641\u06cc \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0645\u06cc\u200c\u06a9\u0646\u062f. \u0627\u06cc\u0646 \u0627\u0633\u0646\u067e\u200c\u0634\u0627\u062a\u200c\u0647\u0627\u060c \u0628\u0647 \u0645\u062f\u0644 \u06a9\u0645\u06a9 \u0645\u06cc\u200c\u06a9\u0646\u0646\u062f \u062a\u0627 \u0628\u0647 \u062a\u062f\u0631\u06cc\u062c \u062f\u0627\u0646\u0634 \u062e\u0648\u062f \u0631\u0627 \u062d\u0641\u0638 \u06a9\u0646\u062f \u0648 \u0627\u0632 \u0641\u0631\u0627\u0645\u0648\u0634\u06cc \u0641\u0627\u062c\u0639\u0647\u200c\u0628\u0627\u0631 \u062c\u0644\u0648\u06af\u06cc\u0631\u06cc \u06a9\u0646\u062f.<\/p>\n<\/p><\/div>\n<h2>\u06cc\u0627\u0641\u062a\u0647\u200c\u0647\u0627\u06cc \u06a9\u0644\u06cc\u062f\u06cc<\/h2>\n<div class=\"section-box\">\n<p>\u0646\u062a\u0627\u06cc\u062c \u0622\u0632\u0645\u0627\u06cc\u0634\u200c\u0647\u0627\u06cc \u0627\u0646\u062c\u0627\u0645 \u0634\u062f\u0647 \u062f\u0631 \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647\u060c \u0646\u0634\u0627\u0646 \u0645\u06cc\u200c\u062f\u0647\u062f \u06a9\u0647 CAP \u0628\u0647 \u0637\u0648\u0631 \u0645\u062f\u0627\u0648\u0645 \u0639\u0645\u0644\u06a9\u0631\u062f \u0628\u0647\u062a\u0631\u06cc \u0646\u0633\u0628\u062a \u0628\u0647 \u0631\u0648\u0634\u200c\u0647\u0627\u06cc \u0647\u0631\u0633 \u0633\u0646\u062a\u06cc \u062f\u0627\u0631\u062f. \u0628\u0647 \u0648\u06cc\u0698\u0647\u060c \u062f\u0631 \u0633\u0646\u0627\u0631\u06cc\u0648\u0647\u0627\u06cc \u0628\u0627 <span class=\"highlight\">\u067e\u0631\u0627\u06a9\u0646\u062f\u06af\u06cc \u0628\u0633\u06cc\u0627\u0631 \u0628\u0627\u0644\u0627<\/span> (\u06cc\u0639\u0646\u06cc \u0632\u0645\u0627\u0646\u06cc \u06a9\u0647 \u062d\u062c\u0645 \u0632\u06cc\u0627\u062f\u06cc \u0627\u0632 \u067e\u0627\u0631\u0627\u0645\u062a\u0631\u0647\u0627\u06cc \u0645\u062f\u0644 \u062d\u0630\u0641 \u0634\u062f\u0647\u200c\u0627\u0646\u062f)\u060c CAP \u0639\u0645\u0644\u06a9\u0631\u062f \u0628\u0633\u06cc\u0627\u0631 \u062e\u0648\u0628\u06cc \u0627\u0632 \u062e\u0648\u062f \u0646\u0634\u0627\u0646 \u0645\u06cc\u200c\u062f\u0647\u062f.<\/p>\n<p>\u0628\u0647 \u0639\u0646\u0648\u0627\u0646 \u0645\u062b\u0627\u0644\u060c \u062f\u0631 \u0648\u0638\u0627\u06cc\u0641 QQP \u0648 MNLI\u060c CAP \u0628\u0627 \u062d\u0641\u0638 \u062a\u0646\u0647\u0627 3 \u062f\u0631\u0635\u062f \u0627\u0632 \u067e\u0627\u0631\u0627\u0645\u062a\u0631\u0647\u0627\u06cc \u0645\u062f\u0644\u060c \u062a\u0648\u0627\u0646\u0633\u062a \u0628\u0647 99.2 \u062f\u0631\u0635\u062f \u0648 96.3 \u062f\u0631\u0635\u062f \u0627\u0632 \u0639\u0645\u0644\u06a9\u0631\u062f \u0627\u0635\u0644\u06cc \u0645\u062f\u0644 BERT \u062f\u0633\u062a \u06cc\u0627\u0628\u062f. \u0627\u06cc\u0646 \u0646\u062a\u0627\u06cc\u062c\u060c \u0646\u0634\u0627\u0646 \u0645\u06cc\u200c\u062f\u0647\u062f \u06a9\u0647 CAP \u06cc\u06a9 \u0631\u0648\u0634 \u0628\u0633\u06cc\u0627\u0631 \u06a9\u0627\u0631\u0622\u0645\u062f \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc \u0628\u0632\u0631\u06af \u0627\u0633\u062a.<\/p>\n<p>\u0639\u0644\u0627\u0648\u0647 \u0628\u0631 \u0627\u06cc\u0646\u060c \u0622\u0632\u0645\u0627\u06cc\u0634\u200c\u0647\u0627\u06cc \u06a9\u0627\u0648\u0634 (probing experiments) \u0646\u0634\u0627\u0646 \u0645\u06cc\u200c\u062f\u0647\u062f \u06a9\u0647 \u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647 \u062a\u0648\u0633\u0637 CAP\u060c \u062a\u0648\u0627\u0646\u0627\u06cc\u06cc <span class=\"highlight\">\u062a\u0639\u0645\u06cc\u0645<\/span> \u0628\u0647\u062a\u0631\u06cc \u062f\u0627\u0631\u062f. \u0627\u06cc\u0646 \u0628\u062f\u0627\u0646 \u0645\u0639\u0646\u0627\u0633\u062a \u06a9\u0647 \u0645\u062f\u0644 CAP\u060c \u0645\u06cc\u200c\u062a\u0648\u0627\u0646\u062f \u0628\u0647 \u0637\u0648\u0631 \u0645\u0648\u062b\u0631\u062a\u0631\u06cc \u0628\u0647 \u062f\u0627\u062f\u0647\u200c\u0647\u0627\u06cc \u062c\u062f\u06cc\u062f \u0648 \u0646\u0627\u0634\u0646\u0627\u062e\u062a\u0647 \u067e\u0627\u0633\u062e \u062f\u0647\u062f.<\/p>\n<\/p><\/div>\n<h2>\u06a9\u0627\u0631\u0628\u0631\u062f\u0647\u0627 \u0648 \u062f\u0633\u062a\u0627\u0648\u0631\u062f\u0647\u0627<\/h2>\n<div class=\"section-box\">\n<p>\u0631\u0648\u0634 CAP\u060c \u06a9\u0627\u0631\u0628\u0631\u062f\u0647\u0627\u06cc \u06af\u0633\u062a\u0631\u062f\u0647\u200c\u0627\u06cc \u062f\u0631 \u0632\u0645\u06cc\u0646\u0647\u200c\u0647\u0627\u06cc \u0645\u062e\u062a\u0644\u0641 NLP \u062f\u0627\u0631\u062f. \u0628\u0647 \u0637\u0648\u0631 \u06a9\u0644\u06cc\u060c \u0627\u06cc\u0646 \u0631\u0648\u0634 \u0645\u06cc\u200c\u062a\u0648\u0627\u0646\u062f \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0647\u0631 \u0646\u0648\u0639 \u0645\u062f\u0644 \u0632\u0628\u0627\u0646\u06cc \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0634\u0648\u062f. \u0628\u0631\u062e\u06cc \u0627\u0632 \u06a9\u0627\u0631\u0628\u0631\u062f\u0647\u0627\u06cc \u062e\u0627\u0635 CAP \u0639\u0628\u0627\u0631\u062a\u0646\u062f \u0627\u0632:<\/p>\n<ul>\n<li><span class=\"highlight\">\u0627\u0633\u062a\u0642\u0631\u0627\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc \u062f\u0631 \u062f\u0633\u062a\u06af\u0627\u0647\u200c\u0647\u0627\u06cc \u062a\u0644\u0641\u0646 \u0647\u0645\u0631\u0627\u0647 \u0648 \u0633\u06cc\u0633\u062a\u0645\u200c\u0647\u0627\u06cc \u062a\u0639\u0628\u06cc\u0647\u200c\u0634\u062f\u0647<\/span>: \u0628\u0627 \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u200c\u0647\u0627\u060c \u0645\u06cc\u200c\u062a\u0648\u0627\u0646 \u0622\u0646\u200c\u0647\u0627 \u0631\u0627 \u0628\u0647 \u0631\u0627\u062d\u062a\u06cc \u062f\u0631 \u062f\u0633\u062a\u06af\u0627\u0647\u200c\u0647\u0627\u06cc \u0628\u0627 \u0645\u062d\u062f\u0648\u062f\u06cc\u062a \u0645\u0646\u0627\u0628\u0639 \u0627\u062c\u0631\u0627 \u06a9\u0631\u062f.<\/li>\n<li><span class=\"highlight\">\u0628\u0647\u0628\u0648\u062f \u0633\u0631\u0639\u062a \u0648 \u06a9\u0627\u0631\u0627\u06cc\u06cc \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc<\/span>: \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u060c \u0633\u0631\u06cc\u0639\u200c\u062a\u0631 \u0648 \u06a9\u0627\u0631\u0622\u0645\u062f\u062a\u0631 \u0627\u0632 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0628\u0632\u0631\u06af \u0647\u0633\u062a\u0646\u062f.<\/li>\n<li><span class=\"highlight\">\u06a9\u0627\u0647\u0634 \u0647\u0632\u06cc\u0646\u0647\u200c\u0647\u0627\u06cc \u0645\u062d\u0627\u0633\u0628\u0627\u062a\u06cc \u0648 \u0645\u0635\u0631\u0641 \u0627\u0646\u0631\u0698\u06cc<\/span>: \u0628\u0627 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u060c \u0645\u06cc\u200c\u062a\u0648\u0627\u0646 \u0647\u0632\u06cc\u0646\u0647\u200c\u0647\u0627\u06cc \u0645\u062d\u0627\u0633\u0628\u0627\u062a\u06cc \u0648 \u0645\u0635\u0631\u0641 \u0627\u0646\u0631\u0698\u06cc \u0631\u0627 \u0628\u0647 \u0637\u0648\u0631 \u0642\u0627\u0628\u0644 \u062a\u0648\u062c\u0647\u06cc \u06a9\u0627\u0647\u0634 \u062f\u0627\u062f.<\/li>\n<\/ul>\n<p>\u062f\u0633\u062a\u0627\u0648\u0631\u062f \u0627\u0635\u0644\u06cc \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647\u060c \u0627\u0631\u0627\u0626\u0647 \u06cc\u06a9 \u0631\u0648\u0634 \u062c\u062f\u06cc\u062f \u0648 \u06a9\u0627\u0631\u0622\u0645\u062f \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0627\u0633\u062a. CAP\u060c \u0628\u0627 \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 \u06cc\u0627\u062f\u06af\u06cc\u0631\u06cc \u062a\u0642\u0627\u0628\u0644\u06cc\u060c \u0645\u06cc\u200c\u062a\u0648\u0627\u0646\u062f \u062a\u0639\u0627\u062f\u0644\u06cc \u0628\u06cc\u0646 \u062d\u0641\u0638 \u062f\u0627\u0646\u0634 \u0639\u0645\u0648\u0645\u06cc \u0648 \u062f\u0627\u0646\u0634 \u062e\u0627\u0635 \u0648\u0638\u06cc\u0641\u0647 \u0628\u0631\u0642\u0631\u0627\u0631 \u06a9\u0646\u062f \u0648 \u062f\u0631 \u0646\u062a\u06cc\u062c\u0647\u060c \u0639\u0645\u0644\u06a9\u0631\u062f \u0645\u062f\u0644 \u0647\u0631\u0633\u200c\u0634\u062f\u0647 \u0631\u0627 \u0628\u0647 \u0637\u0648\u0631 \u0642\u0627\u0628\u0644 \u062a\u0648\u062c\u0647\u06cc \u0628\u0647\u0628\u0648\u062f \u0628\u062e\u0634\u062f.<\/p>\n<\/p><\/div>\n<h2>\u0646\u062a\u06cc\u062c\u0647\u200c\u06af\u06cc\u0631\u06cc<\/h2>\n<div class=\"section-box\">\n<p>\u0645\u0642\u0627\u0644\u0647 &#8220;\u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647&#8221;\u060c \u06cc\u06a9 \u06af\u0627\u0645 \u0645\u0647\u0645 \u062f\u0631 \u0631\u0627\u0633\u062a\u0627\u06cc \u0628\u0647\u0628\u0648\u062f \u06a9\u0627\u0631\u0627\u06cc\u06cc \u0648 \u0642\u0627\u0628\u0644\u06cc\u062a \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0627\u0632 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc \u0628\u0632\u0631\u06af \u0627\u0633\u062a. \u0631\u0648\u0634 CAP\u060c \u06cc\u06a9 \u0631\u0648\u06cc\u06a9\u0631\u062f \u0646\u0648\u0622\u0648\u0631\u0627\u0646\u0647 \u0648 \u06a9\u0627\u0631\u0622\u0645\u062f \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u200c\u0647\u0627 \u0627\u0631\u0627\u0626\u0647 \u0645\u06cc\u200c\u062f\u0647\u062f \u06a9\u0647 \u0645\u06cc\u200c\u062a\u0648\u0627\u0646\u062f \u062f\u0631 \u0632\u0645\u06cc\u0646\u0647\u200c\u0647\u0627\u06cc \u0645\u062e\u062a\u0644\u0641 NLP \u0645\u0648\u0631\u062f \u0627\u0633\u062a\u0641\u0627\u062f\u0647 \u0642\u0631\u0627\u0631 \u06af\u06cc\u0631\u062f. \u0628\u0627 \u062a\u0648\u062c\u0647 \u0628\u0647 \u0627\u0647\u0645\u06cc\u062a \u0631\u0648\u0632\u0627\u0641\u0632\u0648\u0646 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646\u06cc \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647\u060c \u062a\u062d\u0642\u06cc\u0642\u0627\u062a \u062f\u0631 \u0632\u0645\u06cc\u0646\u0647 \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0645\u062f\u0644\u060c \u0627\u0632 \u0627\u0647\u0645\u06cc\u062a \u0648\u06cc\u0698\u0647\u200c\u0627\u06cc \u0628\u0631\u062e\u0648\u0631\u062f\u0627\u0631 \u0627\u0633\u062a \u0648 \u0627\u06cc\u0646 \u0645\u0642\u0627\u0644\u0647\u060c contribution \u0642\u0627\u0628\u0644 \u062a\u0648\u062c\u0647\u06cc \u0628\u0647 \u0627\u06cc\u0646 \u062d\u0648\u0632\u0647 \u0645\u062d\u0633\u0648\u0628 \u0645\u06cc\u200c\u0634\u0648\u062f.<\/p>\n<\/p><\/div>\n<\/article>\n","protected":false},"excerpt":{"rendered":"<p>\ud83d\udcda \u0645\u0642\u0627\u0644\u0647 \u0639\u0644\u0645\u06cc \u0639\u0646\u0648\u0627\u0646 \u0641\u0627\u0631\u0633\u06cc \u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0646\u0648\u06cc\u0633\u0646\u062f\u06af\u0627\u0646 Runxin [&hellip;]<\/p>\n","protected":false},"featured_media":208799,"comment_status":"open","ping_status":"closed","template":"","meta":{"pmpro_default_level":"","site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-gradient":""}}},"product_cat":[137319,137318],"product_tag":[137370,184975,107252,184895,185067,137436,184825,184790,184933,185105,227,139908,185131,184959,184799,184972,185144,184797,184970,185142,63989,184973,185145,40074,185135,184827,185000,184963,184807,184981,83131,185139,184967,91334,184828,185001,184960,185132,182770,184997,184798,184971,185143,184794,184964,185136,91336,184961,185133,184796,184966,185138,91335,184962,185134,137454,184969,185141,130225,184958,185130,184795,184965,185137,57120,184968,185140,184784,184905,185077,83258,184916,185088,352,184985,42885,107,143837,70651,184877,185049,83253,184779,185062,184890,141648,185042,184871,184820,184994,137975,184802,166744,184943,185115,79423,184926,185098,184755,184839,185011,20888,45998,184914,185086,1340,184938,185110,63840,184915,185087,91322,184750,141318,184944,185116,184768,184854,185025,137790,47705,184945,185117,105074,184955,185127,184817,184991,1841,138391,137417,185069,184897,184812,184800,27648,184880,185052,184776,184886,185058,18174,184769,184785,184858,185029,184912,185084,140847,237,184909,185081,41438,184747,140984,184866,185037,1840,184803,184976,184746,184814,42046,184783,184904,185076,184978,140355,140243,184956,185128,46244,184859,185030,137404,184946,185118,184782,184901,185073,171147,103253,75371,185109,184937,141519,184931,185103,58087,139665,184805,66779,138056,185040,184869,172692,184856,185027,184789,184932,185104,184793,184957,185129,184993,137355,184819,18295,185054,184882,61487,184908,185080,184826,91332,185008,91218,184925,185097,46006,185032,184861,137821,184786,184919,185091,46243,185031,184860,137830,138077,63994,184929,185101,184749,184833,185005,42467,184765,184849,185020,184748,184818,184992,8344,184974,184821,184995,184181,184822,11094,137434,1842,11097,184902,185074,127922,185055,184883,142447,164754,184868,185039,83252,102944,57582,184928,185100,62076,86388,184939,185111,184809,184982,184774,184873,185044,179858,45820,61486,184911,185083,1148,184876,185048,11190,184920,185092,184791,184934,185106,166140,239,185,185026,184855,184787,184923,185095,22262,184936,185108,631,184758,185013,184842,100733,184874,185045,184771,184864,185035,139666,139667,184988,3206,184191,184872,185043,138723,184930,185102,53030,184862,185033,20475,184752,184835,184744,142939,184979,137804,184751,184834,185006,111686,184927,185099,143369,184816,184990,138596,170525,184898,185070,184780,184891,185063,166151,184808,139662,184823,115558,175486,57900,184907,185079,142917,184950,185122,137921,140126,172693,185028,184857,138016,1675,28178,142118,184792,184942,185114,1151,184804,184977,184777,184887,185059,184770,184863,185034,184831,185003,184811,184984,1843,184921,185093,184759,185014,184843,1457,184889,185061,18540,185053,184881,166202,184917,185089,83248,143857,184989,184815,184775,184875,185046,91075,184918,185090,371,61340,184951,185123,141839,184952,185124,141549,184829,184788,184924,185096,184764,185019,184848,182919,184947,185119,139364,142504,59312,184903,185075,32420,184870,185041,184761,184845,185016,184801,184847,185018,184763,184760,185015,184844,184762,185017,184846,165216,184954,185126,139158,34383,184941,185113,137431,184773,185038,184867,83259,7558,7515,184913,185085,83251,10810,137452,184900,185072,185002,184830,140367,137928,184806,184980,45189,67770,177,184879,185051,138308,184998,184824,137421,184781,185068,184896,138071,185094,184922,138028,184948,185120,91325,184894,185066,15176,104,170550,185047,184130,184899,185071,83680,185057,184885,49693,184906,185078,87473,184935,185107,141197,184953,185125,175001,184766,184852,185023,184767,184853,185024,184745,173171,185022,184851,184836,166742,185021,184850,184813,184986,184753,13270,166249,184810,184983,5904,184910,185082,184778,185060,184888,173914,184756,184840,185012,143855,185009,184837,139270,138545,184940,185112,35176,184892,185064,106672,140649,184772,184865,185036,18293,184884,185056,139505,184949,185121,63737,49733,184987,184996,185007,91326,139086,185004,184832,15207,83249,184999,173915,184757,184841,184754,185010,184838,31303,184893,185065,49729,49732,31968,184878,185050,128639,91333,137547],"class_list":{"0":"post-224256","1":"product","2":"type-product","3":"status-publish","4":"has-post-thumbnail","6":"product_cat-arxiv","7":"product_cat-137318","8":"product_tag-accuracy","9":"product_tag-achieving-better-plm-compression","10":"product_tag-active-learning","11":"product_tag-active-learning-for-data-annotation","12":"product_tag-active-learning-for-efficient-data-annotation","13":"product_tag-adversarial-attacks","14":"product_tag-adversarial-robustness-of-pruned-models","15":"product_tag-agricultural-ai","16":"product_tag-agricultural-ai-and-smart-farming","17":"product_tag-agricultural-ai-and-smart-farming-with-nlp","18":"product_tag-ai","19":"product_tag-ai-alignment","20":"product_tag-ai-alignment-for-compressed-language-models","21":"product_tag-ai-alignment-for-compressed-models","22":"product_tag-ai-books","23":"product_tag-ai-books-on-deep-learning-compression","24":"product_tag-ai-books-on-deep-learning-model-compression","25":"product_tag-ai-conferences","26":"product_tag-ai-conferences-on-model-compression","27":"product_tag-ai-conferences-on-model-compression-and-efficiency","28":"product_tag-ai-courses","29":"product_tag-ai-courses-on-nlp-and-model-efficiency","30":"product_tag-ai-courses-on-nlp-and-model-efficiency-strategies","31":"product_tag-ai-ethics","32":"product_tag-ai-ethics-in-ai-model-development-lifecycle","33":"product_tag-ai-ethics-in-model-compression","34":"product_tag-ai-ethics-in-model-compression-practices","35":"product_tag-ai-ethics-in-model-development","36":"product_tag-ai-for-edge-devices","37":"product_tag-ai-for-mobile-deployments","38":"product_tag-ai-future","39":"product_tag-ai-future-and-the-role-of-compressed-ai","40":"product_tag-ai-future-and-the-role-of-compressed-models","41":"product_tag-ai-governance","42":"product_tag-ai-governance-for-model-compression","43":"product_tag-ai-governance-for-plm-deployment","44":"product_tag-ai-governance-for-plm-use","45":"product_tag-ai-governance-for-plm-use-cases","46":"product_tag-ai-hardware-acceleration","47":"product_tag-ai-hardware-acceleration-for-plms","48":"product_tag-ai-journals","49":"product_tag-ai-journals-on-efficient-ai","50":"product_tag-ai-journals-on-efficient-ai-and-nlp","51":"product_tag-ai-philosophy","52":"product_tag-ai-philosophy-and-the-nature-of-language-models","53":"product_tag-ai-philosophy-and-the-nature-of-language-understanding","54":"product_tag-ai-policy","55":"product_tag-ai-policy-on-model-compression","56":"product_tag-ai-policy-on-model-compression-standards","57":"product_tag-ai-psychology","58":"product_tag-ai-psychology-and-human-ai-interaction","59":"product_tag-ai-psychology-and-human-ai-interaction-design","60":"product_tag-ai-regulation","61":"product_tag-ai-regulation-for-ai-deployment","62":"product_tag-ai-regulation-for-efficient-ai-deployment","63":"product_tag-ai-research","64":"product_tag-ai-research-in-plm-compression","65":"product_tag-ai-research-in-plm-compression-techniques","66":"product_tag-ai-safety","67":"product_tag-ai-safety-in-plm-deployment","68":"product_tag-ai-safety-in-plm-deployment-strategies","69":"product_tag-ai-sociology","70":"product_tag-ai-sociology-and-impact-on-society","71":"product_tag-ai-sociology-and-societal-impact-of-ai-models","72":"product_tag-ai-trends","73":"product_tag-ai-trends-in-nlp-and-compression","74":"product_tag-ai-trends-in-nlp-and-model-efficiency","75":"product_tag-analytical-models","76":"product_tag-analytical-models-in-nlp","77":"product_tag-analytical-models-in-nlp-and-performance","78":"product_tag-anomaly-detection","79":"product_tag-anomaly-detection-in-text","80":"product_tag-anomaly-detection-in-text-for-robustness","81":"product_tag-applications","82":"product_tag-applying-contrastive-learning-in-pruning","83":"product_tag-architecture-design","84":"product_tag-artificial-intelligence","85":"product_tag-attention-mechanisms","86":"product_tag-audio-processing","87":"product_tag-audio-processing-for-nlp","88":"product_tag-audio-processing-for-nlp-and-efficiency","89":"product_tag-backpropagation","90":"product_tag-batch-learning","91":"product_tag-batch-learning-for-efficient-plm-training","92":"product_tag-batch-learning-for-plms","93":"product_tag-bayesian-methods","94":"product_tag-bayesian-methods-for-nlp-model-inference","95":"product_tag-bayesian-methods-in-nlp","96":"product_tag-benchmarking-compression-methods","97":"product_tag-benchmarking-model-compression-methods","98":"product_tag-benchmarks","99":"product_tag-better-plm-compression","100":"product_tag-bio-inspired-computing","101":"product_tag-bio-inspired-computing-for-nlp","102":"product_tag-bio-inspired-computing-for-nlp-algorithms","103":"product_tag-biometrics","104":"product_tag-biometrics-and-language-processing","105":"product_tag-biometrics-and-language-processing-integration","106":"product_tag-block-sparsity","107":"product_tag-block-sparsity-in-language-models","108":"product_tag-block-sparsity-in-large-language-models","109":"product_tag-chatbots","110":"product_tag-classification","111":"product_tag-classification-tasks-in-nlp","112":"product_tag-classification-tasks-in-nlp-with-compressed-models","113":"product_tag-cloud-computing","114":"product_tag-cloud-computing-for-ai","115":"product_tag-cloud-computing-for-large-scale-ai-training","116":"product_tag-clustering","117":"product_tag-clustering-text-data","118":"product_tag-clustering-text-data-for-efficiency","119":"product_tag-cnns","120":"product_tag-coarse-grained-pruning","121":"product_tag-cognitive-computing","122":"product_tag-cognitive-computing-and-language-understanding","123":"product_tag-cognitive-computing-and-language-understanding-systems","124":"product_tag-compressed-sensing","125":"product_tag-compressed-sensing-in-nlp","126":"product_tag-compressed-sensing-principles-in-nlp","127":"product_tag-computational-cost","128":"product_tag-computational-linguistics","129":"product_tag-computational-linguistics-and-model-compression","130":"product_tag-computational-linguistics-and-model-compression-research","131":"product_tag-computational-social-science","132":"product_tag-computational-social-science-and-nlp","133":"product_tag-computational-social-science-and-nlp-applications","134":"product_tag-computationally-efficient-models","135":"product_tag-computationally-efficient-models-for-nlp","136":"product_tag-computer-vision","137":"product_tag-continual-learning","138":"product_tag-contrastive-learning","139":"product_tag-contrastive-learning-for-improved-plms","140":"product_tag-contrastive-learning-for-plms","141":"product_tag-contrastive-learning-in-pruning","142":"product_tag-contrastive-pruning-for-plm-compression","143":"product_tag-control-systems","144":"product_tag-control-systems-for-ai-agents","145":"product_tag-control-systems-for-ai-agents-and-efficiency","146":"product_tag-convex-optimization","147":"product_tag-convex-optimization-in-machine-learning","148":"product_tag-convex-optimization-in-machine-learning-models","149":"product_tag-convolutional-neural-networks","150":"product_tag-correlation","151":"product_tag-correlation-analysis","152":"product_tag-correlation-analysis-in-model-pruning","153":"product_tag-correlation-analysis-in-model-pruning-decisions","154":"product_tag-correlation-analysis-in-text-data","155":"product_tag-correlation-analysis-in-text-data-for-pruning","156":"product_tag-cross-modal-learning","157":"product_tag-data-analysis","158":"product_tag-data-analysis-for-language","159":"product_tag-data-analysis-for-nlp-model-efficiency","160":"product_tag-data-augmentation","161":"product_tag-dataset-curation","162":"product_tag-decision-trees","163":"product_tag-decision-trees-for-nlp","164":"product_tag-decision-trees-for-nlp-and-efficiency","165":"product_tag-deep-learning","166":"product_tag-deep-learning-model-compression","167":"product_tag-deep-learning-model-compression-techniques","168":"product_tag-dense-models","169":"product_tag-dense-to-sparse-transition","170":"product_tag-deployment","171":"product_tag-descriptive-models","172":"product_tag-descriptive-models-for-text-analysis","173":"product_tag-descriptive-models-for-text-analysis-and-compression","174":"product_tag-developing-efficient-ai-models","175":"product_tag-dialogue-systems","176":"product_tag-digital-humanities","177":"product_tag-digital-humanities-and-text-analysis","178":"product_tag-digital-humanities-and-text-analysis-tools","179":"product_tag-dimensionality-reduction","180":"product_tag-dimensionality-reduction-in-nlp","181":"product_tag-dimensionality-reduction-techniques-in-nlp","182":"product_tag-discourse-analysis","183":"product_tag-discourse-analysis-and-pruning","184":"product_tag-discourse-analysis-and-pruning-strategies","185":"product_tag-discriminative-learning","186":"product_tag-discriminative-learning-for-classification","187":"product_tag-discriminative-learning-for-nlp-classification-and-efficiency","188":"product_tag-distributed-learning","189":"product_tag-edge-ai","190":"product_tag-edge-computing","191":"product_tag-edge-computing-for-nlp-model-deployment","192":"product_tag-edge-computing-for-nlp-models","193":"product_tag-educational-ai","194":"product_tag-educational-ai-and-language-learning","195":"product_tag-educational-ai-and-language-learning-technologies","196":"product_tag-efficiency","197":"product_tag-efficient-ai","198":"product_tag-efficient-ai-models","199":"product_tag-embeddings","200":"product_tag-ensemble-methods","201":"product_tag-ensemble-methods-for-robust-nlp-models","202":"product_tag-ensemble-methods-in-nlp","203":"product_tag-entropy","204":"product_tag-entropy-in-plm-pruning","205":"product_tag-entropy-measures-in-plm-pruning","206":"product_tag-environmental-ai","207":"product_tag-environmental-ai-and-nlp","208":"product_tag-environmental-ai-and-nlp-for-monitoring","209":"product_tag-ethics-of-ai","210":"product_tag-ethics-of-ai-in-model-compression","211":"product_tag-ethics-of-ai-in-model-compression-practices","212":"product_tag-evaluating-compressed-ai-models","213":"product_tag-evaluation-metrics","214":"product_tag-evaluation-of-compressed-models","215":"product_tag-evolutionary-algorithms","216":"product_tag-evolutionary-algorithms-for-hyperparameter-optimization","217":"product_tag-evolutionary-algorithms-for-hyperparameter-tuning","218":"product_tag-experimentation","219":"product_tag-experimentation-in-nlp","220":"product_tag-experimentation-in-nlp-model-compression","221":"product_tag-explainability-of-pruning-decisions","222":"product_tag-explainable-ai","223":"product_tag-exploring-sparse-patterns-in-language-models","224":"product_tag-facial-recognition","225":"product_tag-facial-recognition-in-nlp-applications","226":"product_tag-facial-recognition-in-nlp-applications-and-deployment","227":"product_tag-feature-engineering","228":"product_tag-feature-engineering-for-efficient-nlp","229":"product_tag-feature-engineering-for-language-models","230":"product_tag-feature-extraction-2","231":"product_tag-feature-matching","232":"product_tag-feature-matching-in-nlp","233":"product_tag-feature-matching-in-nlp-and-multimodal-ai","234":"product_tag-feature-selection","235":"product_tag-feature-selection-for-language-models","236":"product_tag-feature-selection-for-plms","237":"product_tag-federated-learning","238":"product_tag-few-shot-learning","239":"product_tag-financial-ai","240":"product_tag-financial-ai-and-nlp","241":"product_tag-financial-ai-and-nlp-for-business","242":"product_tag-fine-grained-pruning","243":"product_tag-fine-grained-vs-coarse-grained-pruning","244":"product_tag-fine-grained-vs-coarse-grained-pruning-in-nlp","245":"product_tag-fine-tuning","246":"product_tag-fine-tuning-after-pruning","247":"product_tag-fine-tuning-after-pruning-plms","248":"product_tag-fine-tuning-after-pruning-plms-for-performance","249":"product_tag-fine-tuning-strategies","250":"product_tag-fine-tuning-strategies-for-pruned-models","251":"product_tag-fine-tuning-strategies-for-pruned-plms","252":"product_tag-foundation-models","253":"product_tag-from-dense-to-sparse-language-models","254":"product_tag-frontiers-in-ai-compression","255":"product_tag-frontiers-in-ai-model-compression-research","256":"product_tag-future-directions","257":"product_tag-future-of-nlp-models","258":"product_tag-gans","259":"product_tag-generalization","260":"product_tag-generative-adversarial-networks","261":"product_tag-generative-models","262":"product_tag-generative-models-for-text","263":"product_tag-generative-models-for-text-and-compression","264":"product_tag-genetic-algorithms","265":"product_tag-genetic-algorithms-for-ai-model-design","266":"product_tag-genetic-algorithms-for-model-design","267":"product_tag-gnns","268":"product_tag-gradient-boosting","269":"product_tag-gradient-boosting-for-nlp","270":"product_tag-gradient-boosting-for-nlp-applications","271":"product_tag-gradient-descent","272":"product_tag-graph-neural-networks","273":"product_tag-healthcare-ai","274":"product_tag-healthcare-ai-and-plm-compression","275":"product_tag-healthcare-ai-and-plm-compression-strategies","276":"product_tag-hidden-markov-models","277":"product_tag-high-performance-computing","278":"product_tag-high-performance-computing-for-ai-research","279":"product_tag-high-performance-computing-for-ai-research-and-nlp","280":"product_tag-high-performance-nlp","281":"product_tag-high-performance-nlp-applications","282":"product_tag-hmms","283":"product_tag-hmms-for-sequence-labeling","284":"product_tag-hmms-for-sequence-labeling-and-pruning","285":"product_tag-hpc","286":"product_tag-hyperparameter-tuning","287":"product_tag-hypothesis-testing","288":"product_tag-hypothesis-testing-in-nlp","289":"product_tag-hypothesis-testing-in-nlp-model-evaluation","290":"product_tag-image-processing","291":"product_tag-image-processing-in-multimodal-nlp","292":"product_tag-image-processing-in-multimodal-nlp-and-compression","293":"product_tag-image-recognition","294":"product_tag-image-recognition-for-multimodal-ai","295":"product_tag-image-recognition-in-multimodal-ai-and-compression","296":"product_tag-industrial-ai","297":"product_tag-industrial-ai-and-nlp-applications","298":"product_tag-industrial-ai-and-nlp-applications-in-manufacturing","299":"product_tag-inference-speed","300":"product_tag-information-retrieval","301":"product_tag-information-theory","302":"product_tag-information-theory-applied-to-model-compression","303":"product_tag-information-theory-in-model-compression","304":"product_tag-instance-segmentation","305":"product_tag-instance-segmentation-for-nlp","306":"product_tag-instance-segmentation-in-nlp-tasks","307":"product_tag-internet-of-things","308":"product_tag-internet-of-things-and-nlp","309":"product_tag-internet-of-things-and-nlp-for-data-analysis","310":"product_tag-iot","311":"product_tag-iterative-pruning","312":"product_tag-iterative-pruning-for-optimal-plm-compression","313":"product_tag-iterative-pruning-for-plms","314":"product_tag-kalman-filters","315":"product_tag-kalman-filters-in-nlp","316":"product_tag-kalman-filters-in-nlp-applications","317":"product_tag-kernel-methods","318":"product_tag-kernel-methods-in-nlp","319":"product_tag-kernel-methods-in-nlp-model-compression","320":"product_tag-knowledge-distillation","321":"product_tag-language-model-optimization","322":"product_tag-language-model-optimization-techniques","323":"product_tag-large-language-models","324":"product_tag-latent-variable-models","325":"product_tag-latent-variable-models-for-text","326":"product_tag-latent-variable-models-for-text-representation","327":"product_tag-legal-ai","328":"product_tag-legal-ai-and-text-analysis","329":"product_tag-legal-ai-and-text-analysis-for-efficiency","330":"product_tag-linear-models","331":"product_tag-linear-models-for-text-analysis","332":"product_tag-linear-models-in-text-analysis-and-compression","333":"product_tag-llms","334":"product_tag-lottery-ticket-hypothesis","335":"product_tag-lottery-ticket-hypothesis-in-plms","336":"product_tag-low-rank-factorization","337":"product_tag-low-resource-nlp","338":"product_tag-low-resource-nlp-solutions","339":"product_tag-machine-translation","340":"product_tag-magnitude-pruning","341":"product_tag-magnitude-pruning-for-plms","342":"product_tag-magnitude-pruning-for-pre-trained-language-models","343":"product_tag-medical-imaging","344":"product_tag-medical-imaging-and-nlp","345":"product_tag-medical-imaging-and-nlp-applications","346":"product_tag-memory-footprint","347":"product_tag-memory-saving-ai","348":"product_tag-memory-saving-ai-solutions","349":"product_tag-meta-learning","350":"product_tag-metric-learning","351":"product_tag-metric-learning-for-text-embeddings","352":"product_tag-metric-learning-for-text-embeddings-and-pruning","353":"product_tag-mini-batch-learning","354":"product_tag-mini-batch-learning-for-plms","355":"product_tag-mini-batch-learning-for-scalable-nlp","356":"product_tag-mobile-ai","357":"product_tag-mobile-ai-deployments","358":"product_tag-model-compression","359":"product_tag-model-interpretability-in-compressed-models","360":"product_tag-model-optimization","361":"product_tag-model-scaling","362":"product_tag-modeling","363":"product_tag-modeling-language-behavior","364":"product_tag-modeling-language-behavior-with-compressed-models","365":"product_tag-morphology","366":"product_tag-morphology-and-word-structure","367":"product_tag-morphology-and-word-structure-analysis","368":"product_tag-multi-task-learning","369":"product_tag-multimodal-learning","370":"product_tag-mutual-information","371":"product_tag-mutual-information-for-effective-pruning","372":"product_tag-mutual-information-for-pruning","373":"product_tag-named-entity-recognition","374":"product_tag-natural-language-processing","375":"product_tag-ner","376":"product_tag-network-pruning","377":"product_tag-neuromorphic-computing","378":"product_tag-neuromorphic-computing-for-ai","379":"product_tag-neuromorphic-computing-for-efficient-ai","380":"product_tag-nlp","381":"product_tag-nlp-model-compression","382":"product_tag-nlp-model-compression-strategies","383":"product_tag-non-convex-optimization","384":"product_tag-non-convex-optimization-in-deep-learning","385":"product_tag-non-convex-optimization-in-deep-learning-training","386":"product_tag-non-linear-models","387":"product_tag-non-linear-models-for-nlp","388":"product_tag-non-linear-models-in-nlp-and-efficiency","389":"product_tag-novel-applications-of-compressed-models","390":"product_tag-novel-applications-of-compressed-plms","391":"product_tag-novel-pruning-techniques","392":"product_tag-novel-pruning-techniques-for-plms","393":"product_tag-object-detection","394":"product_tag-object-detection-for-multimodal-ai","395":"product_tag-object-detection-in-multimodal-ai-and-efficiency","396":"product_tag-one-shot-pruning","397":"product_tag-one-shot-pruning-for-fast-deployment","398":"product_tag-one-shot-pruning-for-plms","399":"product_tag-online-learning","400":"product_tag-online-learning-for-dynamic-nlp","401":"product_tag-online-learning-for-dynamic-nlp-systems","402":"product_tag-optimization-algorithms","403":"product_tag-optimization-algorithms-for-ai-model-training","404":"product_tag-optimization-algorithms-for-model-training","405":"product_tag-outlier-detection","406":"product_tag-outlier-detection-in-language-data","407":"product_tag-outlier-detection-in-language-data-and-compression","408":"product_tag-overfitting","409":"product_tag-parameter-pruning","410":"product_tag-parameter-reduction-strategies-for-plms","411":"product_tag-parameter-reduction-techniques","412":"product_tag-particle-filters","413":"product_tag-particle-filters-for-language","414":"product_tag-particle-filters-for-language-modeling","415":"product_tag-pattern-recognition","416":"product_tag-pattern-recognition-in-text","417":"product_tag-pattern-recognition-in-text-for-nlp-applications","418":"product_tag-performance","419":"product_tag-phonetics","420":"product_tag-phonetics-and-speech-processing","421":"product_tag-phonetics-and-speech-processing-efficiency","422":"product_tag-phonology","423":"product_tag-phonology-and-sound-systems","424":"product_tag-phonology-and-sound-systems-analysis","425":"product_tag-plms","426":"product_tag-policy-implications-of-plm-compression","427":"product_tag-pose-estimation","428":"product_tag-pose-estimation-in-human-ai-interaction","429":"product_tag-pose-estimation-in-human-ai-interaction-and-efficiency","430":"product_tag-post-training-pruning","431":"product_tag-post-training-pruning-of-language-models","432":"product_tag-post-training-pruning-of-plms","433":"product_tag-pragmatics","434":"product_tag-pragmatics-and-language-understanding","435":"product_tag-pragmatics-and-language-understanding-in-compressed-models","436":"product_tag-pre-trained-language-models","437":"product_tag-pre-training-objectives","438":"product_tag-predictive-models","439":"product_tag-predictive-models-for-language-tasks","440":"product_tag-predictive-models-for-language-tasks-and-efficiency","441":"product_tag-probabilistic-models","442":"product_tag-probabilistic-models-for-language","443":"product_tag-probabilistic-models-in-language-and-compression","444":"product_tag-pruning-criteria","445":"product_tag-pruning-criteria-in-nlp","446":"product_tag-pruning-criteria-in-nlp-research","447":"product_tag-pruning-dense-to-sparse-models","448":"product_tag-pruning-during-plm-training","449":"product_tag-pruning-during-plm-training-phases","450":"product_tag-pruning-during-training","451":"product_tag-pruning-masks","452":"product_tag-pruning-masks-for-efficient-language-models","453":"product_tag-pruning-masks-for-language-models","454":"product_tag-pruning-schedules","455":"product_tag-pruning-schedules-for-plm-optimization","456":"product_tag-pruning-schedules-for-plms","457":"product_tag-psycholinguistics","458":"product_tag-psycholinguistics-and-language-acquisition","459":"product_tag-psycholinguistics-and-language-acquisition-models","460":"product_tag-quantization","461":"product_tag-quantum-computing","462":"product_tag-quantum-computing-for-ai","463":"product_tag-quantum-computing-for-ai-and-language-models","464":"product_tag-question-answering","465":"product_tag-random-forests","466":"product_tag-random-forests-for-language-modeling","467":"product_tag-random-forests-for-nlp","468":"product_tag-recommendation-systems","469":"product_tag-recurrent-neural-networks","470":"product_tag-regression-analysis","471":"product_tag-regression-analysis-for-nlp","472":"product_tag-regression-analysis-for-nlp-model-prediction","473":"product_tag-regularization","474":"product_tag-reinforcement-learning","475":"product_tag-representation-learning","476":"product_tag-representation-learning-in-nlp","477":"product_tag-representation-learning-in-nlp-for-compression","478":"product_tag-research-opportunities-in-ai-model-compression","479":"product_tag-research-opportunities-in-model-compression","480":"product_tag-research-trends","481":"product_tag-resource-efficiency","482":"product_tag-resource-efficient-ai","483":"product_tag-resource-efficient-ai-for-edge-devices","484":"product_tag-responsible-ai","485":"product_tag-rnns","486":"product_tag-robotics","487":"product_tag-robotics-and-nlp","488":"product_tag-robotics-and-nlp-integration","489":"product_tag-robustness","490":"product_tag-robustness-of-compressed-ai-models","491":"product_tag-robustness-of-compressed-models","492":"product_tag-self-supervised-learning","493":"product_tag-self-supervised-learning-objectives","494":"product_tag-self-supervised-learning-objectives-for-plm-optimization","495":"product_tag-self-supervised-learning-objectives-for-plms","496":"product_tag-semantic-segmentation","497":"product_tag-semantic-segmentation-for-efficient-nlp","498":"product_tag-semantic-segmentation-for-multimodal-ai","499":"product_tag-semantics","500":"product_tag-semantics-and-meaning-representation","501":"product_tag-semantics-and-meaning-representation-in-nlp","502":"product_tag-semi-supervised-learning","503":"product_tag-semi-supervised-learning-for-nlp","504":"product_tag-semi-supervised-learning-for-nlp-data","505":"product_tag-sentiment-analysis","506":"product_tag-signal-processing","507":"product_tag-signal-processing-for-nlp","508":"product_tag-signal-processing-techniques-for-nlp","509":"product_tag-similarity-learning","510":"product_tag-similarity-learning-for-text","511":"product_tag-similarity-learning-for-text-representation","512":"product_tag-simulated-annealing","513":"product_tag-simulated-annealing-for-model-tuning","514":"product_tag-simulated-annealing-for-optimization","515":"product_tag-simulation","516":"product_tag-simulation-of-language-processes","517":"product_tag-simulation-of-language-processes-for-understanding","518":"product_tag-smart-cities","519":"product_tag-smart-cities-and-ai-integration","520":"product_tag-smart-cities-and-ai-integration-with-nlp","521":"product_tag-sociolinguistics","522":"product_tag-sociolinguistics-and-language-variation","523":"product_tag-sociolinguistics-and-language-variation-in-nlp","524":"product_tag-sota","525":"product_tag-sparse-autoencoders","526":"product_tag-sparse-autoencoders-for-text","527":"product_tag-sparse-autoencoders-for-text-data","528":"product_tag-sparse-coding","529":"product_tag-sparse-coding-for-language","530":"product_tag-sparse-coding-techniques-for-language","531":"product_tag-sparse-models","532":"product_tag-sparse-networks","533":"product_tag-sparse-networks-for-language-modeling","534":"product_tag-sparse-networks-for-language-models","535":"product_tag-sparse-patterns-in-language-models","536":"product_tag-sparse-representation","537":"product_tag-sparse-representation-in-natural-language-processing","538":"product_tag-sparse-representation-in-nlp","539":"product_tag-sparse-representation-learning","540":"product_tag-sparse-representation-learning-for-language","541":"product_tag-sparsity-patterns","542":"product_tag-speech-recognition","543":"product_tag-state-of-the-art-2","544":"product_tag-state-of-the-art-plm-compression","545":"product_tag-state-of-the-art-plm-compression-research","546":"product_tag-statistical-analysis","547":"product_tag-statistical-analysis-of-nlp-data","548":"product_tag-statistical-analysis-of-nlp-performance-data","549":"product_tag-stochastic-optimization","550":"product_tag-stochastic-optimization-for-large-plms","551":"product_tag-stochastic-optimization-for-plms","552":"product_tag-structured-pruning","553":"product_tag-structured-pruning-methods","554":"product_tag-structured-pruning-methods-for-nlp","555":"product_tag-structured-pruning-methods-for-nlp-applications","556":"product_tag-structured-sparsity","557":"product_tag-structured-sparsity-for-efficient-plms","558":"product_tag-structured-sparsity-for-plms","559":"product_tag-summarization","560":"product_tag-supercomputing","561":"product_tag-supercomputing-for-large-scale-nlp","562":"product_tag-supercomputing-for-large-scale-nlp-model-training","563":"product_tag-supervised-learning","564":"product_tag-supervised-learning-in-nlp","565":"product_tag-supervised-learning-in-nlp-model-compression","566":"product_tag-support-vector-machines","567":"product_tag-sustainable-ai","568":"product_tag-svms","569":"product_tag-svms-for-text-classification","570":"product_tag-svms-for-text-classification-and-pruning","571":"product_tag-swarm-intelligence","572":"product_tag-swarm-intelligence-for-ai","573":"product_tag-swarm-intelligence-in-ai-optimization","574":"product_tag-syntax","575":"product_tag-syntax-and-language-structure","576":"product_tag-syntax-and-language-structure-analysis","577":"product_tag-text-classification","578":"product_tag-text-generation","579":"product_tag-the-dense-to-sparse-transition-in-models","580":"product_tag-the-future-of-nlp-models-and-efficiency","581":"product_tag-the-lottery-ticket-hypothesis-in-plm-compression","582":"product_tag-transfer-learning","583":"product_tag-transferability","584":"product_tag-transferability-of-pruned-language-models","585":"product_tag-transferability-of-pruned-models","586":"product_tag-transformers","587":"product_tag-underfitting","588":"product_tag-understanding-pruning-decisions","589":"product_tag-unstructured-pruning","590":"product_tag-unstructured-pruning-methods","591":"product_tag-unstructured-pruning-methods-for-plms","592":"product_tag-unstructured-sparsity","593":"product_tag-unstructured-sparsity-for-plm-compression","594":"product_tag-unstructured-sparsity-for-plms","595":"product_tag-unsupervised-learning","596":"product_tag-unsupervised-learning-for-language","597":"product_tag-unsupervised-learning-for-language-model-efficiency","598":"product_tag-vaes","599":"product_tag-variational-autoencoders","600":"product_tag-video-processing","601":"product_tag-video-processing-in-multimodal-nlp","602":"product_tag-video-processing-in-multimodal-nlp-and-deployment","603":"product_tag-virtual-assistants","604":"product_tag-xai","605":"product_tag-zero-shot-learning","606":"pmpro-has-access","607":"desktop-align-left","608":"tablet-align-left","609":"mobile-align-left","611":"first","612":"instock","613":"downloadable","614":"virtual","615":"purchasable","616":"product-type-simple"},"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v22.0 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>\u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0628\u0647 \u0647\u0645\u0631\u0627\u0647 PDF \u0641\u0627\u0631\u0633\u06cc + \u067e\u0627\u062f\u06a9\u0633\u062a \u0635\u0648\u062a\u06cc \u0641\u0627\u0631\u0633\u06cc + \u0648\u06cc\u062f\u06cc\u0648 \u0622\u0645\u0648\u0632\u0634\u06cc \u0641\u0627\u0631\u0633\u06cc - \u0641\u0631\u0648\u0634\u06af\u0627\u0647 \u0627\u06a9\u0633\u067e\u0631\u0633<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/express24.ir\/d\/product\/2112-07198\/\" \/>\n<meta property=\"og:locale\" content=\"fa_IR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"\u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0628\u0647 \u0647\u0645\u0631\u0627\u0647 PDF \u0641\u0627\u0631\u0633\u06cc + \u067e\u0627\u062f\u06a9\u0633\u062a \u0635\u0648\u062a\u06cc \u0641\u0627\u0631\u0633\u06cc + \u0648\u06cc\u062f\u06cc\u0648 \u0622\u0645\u0648\u0632\u0634\u06cc \u0641\u0627\u0631\u0633\u06cc - \u0641\u0631\u0648\u0634\u06af\u0627\u0647 \u0627\u06a9\u0633\u067e\u0631\u0633\" \/>\n<meta property=\"og:description\" content=\"\ud83d\udcda \u0645\u0642\u0627\u0644\u0647 \u0639\u0644\u0645\u06cc \u0639\u0646\u0648\u0627\u0646 \u0641\u0627\u0631\u0633\u06cc \u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0646\u0648\u06cc\u0633\u0646\u062f\u06af\u0627\u0646 Runxin [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/express24.ir\/d\/product\/2112-07198\/\" \/>\n<meta property=\"og:site_name\" content=\"\u0641\u0631\u0648\u0634\u06af\u0627\u0647 \u0627\u06a9\u0633\u067e\u0631\u0633\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u0632\u0645\u0627\u0646 \u062a\u0642\u0631\u06cc\u0628\u06cc \u0628\u0631\u0627\u06cc \u062e\u0648\u0627\u0646\u062f\u0646\" \/>\n\t<meta name=\"twitter:data1\" content=\"1 \u062f\u0642\u06cc\u0642\u0647\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/express24.ir\/d\/product\/2112-07198\/\",\"url\":\"https:\/\/express24.ir\/d\/product\/2112-07198\/\",\"name\":\"\u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0628\u0647 \u0647\u0645\u0631\u0627\u0647 PDF \u0641\u0627\u0631\u0633\u06cc + \u067e\u0627\u062f\u06a9\u0633\u062a \u0635\u0648\u062a\u06cc \u0641\u0627\u0631\u0633\u06cc + \u0648\u06cc\u062f\u06cc\u0648 \u0622\u0645\u0648\u0632\u0634\u06cc \u0641\u0627\u0631\u0633\u06cc - \u0641\u0631\u0648\u0634\u06af\u0627\u0647 \u0627\u06a9\u0633\u067e\u0631\u0633\",\"isPartOf\":{\"@id\":\"https:\/\/express24.ir\/d\/#website\"},\"datePublished\":\"2025-11-02T20:19:34+00:00\",\"dateModified\":\"2025-11-02T20:19:34+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/express24.ir\/d\/product\/2112-07198\/#breadcrumb\"},\"inLanguage\":\"fa-IR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/express24.ir\/d\/product\/2112-07198\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/express24.ir\/d\/product\/2112-07198\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u062e\u0627\u0646\u0647\",\"item\":\"https:\/\/express24.ir\/d\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"\u0641\u0631\u0648\u0634\u06af\u0627\u0647\",\"item\":\"https:\/\/express24.ir\/d\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"\u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0628\u0647 \u0647\u0645\u0631\u0627\u0647 PDF \u0641\u0627\u0631\u0633\u06cc + \u067e\u0627\u062f\u06a9\u0633\u062a \u0635\u0648\u062a\u06cc \u0641\u0627\u0631\u0633\u06cc + \u0648\u06cc\u062f\u06cc\u0648 \u0622\u0645\u0648\u0632\u0634\u06cc \u0641\u0627\u0631\u0633\u06cc\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/express24.ir\/d\/#website\",\"url\":\"https:\/\/express24.ir\/d\/\",\"name\":\"\u0641\u0631\u0648\u0634\u06af\u0627\u0647 \u0627\u06a9\u0633\u067e\u0631\u0633\",\"description\":\"\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/express24.ir\/d\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"fa-IR\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"\u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0628\u0647 \u0647\u0645\u0631\u0627\u0647 PDF \u0641\u0627\u0631\u0633\u06cc + \u067e\u0627\u062f\u06a9\u0633\u062a \u0635\u0648\u062a\u06cc \u0641\u0627\u0631\u0633\u06cc + \u0648\u06cc\u062f\u06cc\u0648 \u0622\u0645\u0648\u0632\u0634\u06cc \u0641\u0627\u0631\u0633\u06cc - \u0641\u0631\u0648\u0634\u06af\u0627\u0647 \u0627\u06a9\u0633\u067e\u0631\u0633","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/express24.ir\/d\/product\/2112-07198\/","og_locale":"fa_IR","og_type":"article","og_title":"\u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0628\u0647 \u0647\u0645\u0631\u0627\u0647 PDF \u0641\u0627\u0631\u0633\u06cc + \u067e\u0627\u062f\u06a9\u0633\u062a \u0635\u0648\u062a\u06cc \u0641\u0627\u0631\u0633\u06cc + \u0648\u06cc\u062f\u06cc\u0648 \u0622\u0645\u0648\u0632\u0634\u06cc \u0641\u0627\u0631\u0633\u06cc - \u0641\u0631\u0648\u0634\u06af\u0627\u0647 \u0627\u06a9\u0633\u067e\u0631\u0633","og_description":"\ud83d\udcda \u0645\u0642\u0627\u0644\u0647 \u0639\u0644\u0645\u06cc \u0639\u0646\u0648\u0627\u0646 \u0641\u0627\u0631\u0633\u06cc \u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0646\u0648\u06cc\u0633\u0646\u062f\u06af\u0627\u0646 Runxin [&hellip;]","og_url":"https:\/\/express24.ir\/d\/product\/2112-07198\/","og_site_name":"\u0641\u0631\u0648\u0634\u06af\u0627\u0647 \u0627\u06a9\u0633\u067e\u0631\u0633","twitter_card":"summary_large_image","twitter_misc":{"\u0632\u0645\u0627\u0646 \u062a\u0642\u0631\u06cc\u0628\u06cc \u0628\u0631\u0627\u06cc \u062e\u0648\u0627\u0646\u062f\u0646":"1 \u062f\u0642\u06cc\u0642\u0647"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/express24.ir\/d\/product\/2112-07198\/","url":"https:\/\/express24.ir\/d\/product\/2112-07198\/","name":"\u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0628\u0647 \u0647\u0645\u0631\u0627\u0647 PDF \u0641\u0627\u0631\u0633\u06cc + \u067e\u0627\u062f\u06a9\u0633\u062a \u0635\u0648\u062a\u06cc \u0641\u0627\u0631\u0633\u06cc + \u0648\u06cc\u062f\u06cc\u0648 \u0622\u0645\u0648\u0632\u0634\u06cc \u0641\u0627\u0631\u0633\u06cc - \u0641\u0631\u0648\u0634\u06af\u0627\u0647 \u0627\u06a9\u0633\u067e\u0631\u0633","isPartOf":{"@id":"https:\/\/express24.ir\/d\/#website"},"datePublished":"2025-11-02T20:19:34+00:00","dateModified":"2025-11-02T20:19:34+00:00","breadcrumb":{"@id":"https:\/\/express24.ir\/d\/product\/2112-07198\/#breadcrumb"},"inLanguage":"fa-IR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/express24.ir\/d\/product\/2112-07198\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/express24.ir\/d\/product\/2112-07198\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u062e\u0627\u0646\u0647","item":"https:\/\/express24.ir\/d\/"},{"@type":"ListItem","position":2,"name":"\u0641\u0631\u0648\u0634\u06af\u0627\u0647","item":"https:\/\/express24.ir\/d\/"},{"@type":"ListItem","position":3,"name":"\u0645\u0642\u0627\u0644\u0647 \u0627\u0632 \u0645\u062a\u0631\u0627\u06a9\u0645 \u0628\u0647 \u062a\u064f\u0646\u064f\u06a9: \u0647\u0631\u0633 \u062a\u0642\u0627\u0628\u0644\u06cc \u0628\u0631\u0627\u06cc \u0641\u0634\u0631\u062f\u0647\u200c\u0633\u0627\u0632\u06cc \u0628\u0647\u062a\u0631 \u0645\u062f\u0644\u200c\u0647\u0627\u06cc \u0632\u0628\u0627\u0646 \u067e\u06cc\u0634\u200c\u0622\u0645\u0648\u062e\u062a\u0647 \u0628\u0647 \u0647\u0645\u0631\u0627\u0647 PDF \u0641\u0627\u0631\u0633\u06cc + \u067e\u0627\u062f\u06a9\u0633\u062a \u0635\u0648\u062a\u06cc \u0641\u0627\u0631\u0633\u06cc + \u0648\u06cc\u062f\u06cc\u0648 \u0622\u0645\u0648\u0632\u0634\u06cc \u0641\u0627\u0631\u0633\u06cc"}]},{"@type":"WebSite","@id":"https:\/\/express24.ir\/d\/#website","url":"https:\/\/express24.ir\/d\/","name":"\u0641\u0631\u0648\u0634\u06af\u0627\u0647 \u0627\u06a9\u0633\u067e\u0631\u0633","description":"","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/express24.ir\/d\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"fa-IR"}]}},"_links":{"self":[{"href":"https:\/\/express24.ir\/d\/wp-json\/wp\/v2\/product\/224256","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/express24.ir\/d\/wp-json\/wp\/v2\/product"}],"about":[{"href":"https:\/\/express24.ir\/d\/wp-json\/wp\/v2\/types\/product"}],"replies":[{"embeddable":true,"href":"https:\/\/express24.ir\/d\/wp-json\/wp\/v2\/comments?post=224256"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/express24.ir\/d\/wp-json\/wp\/v2\/media\/208799"}],"wp:attachment":[{"href":"https:\/\/express24.ir\/d\/wp-json\/wp\/v2\/media?parent=224256"}],"wp:term":[{"taxonomy":"product_cat","embeddable":true,"href":"https:\/\/express24.ir\/d\/wp-json\/wp\/v2\/product_cat?post=224256"},{"taxonomy":"product_tag","embeddable":true,"href":"https:\/\/express24.ir\/d\/wp-json\/wp\/v2\/product_tag?post=224256"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}