{"id":3015,"date":"2026-04-28T09:03:41","date_gmt":"2026-04-28T00:03:41","guid":{"rendered":"https:\/\/news.sktelecom.com\/en\/?p=3015"},"modified":"2026-04-28T09:03:41","modified_gmt":"2026-04-28T00:03:41","slug":"sk-telecom-presents-ai-reasoning-model-at-iclr-2026-to-identify-customers-true-preferences","status":"publish","type":"post","link":"https:\/\/news.sktelecom.com\/en\/3015","title":{"rendered":"SK Telecom Presents AI Reasoning Model at ICLR 2026 to Identify Customers\u2019 True Preferences"},"content":{"rendered":"

\u00b7 Paper on AI recommendation model identifying customers\u2019 true preferences presented at ICLR 2026.<\/strong>
\n\u00b7 Proprietary Conflict-Aware Direct Preference Optimization (C-APO) technology improves recommendation performance by distinguishing customers\u2019 revealed and coherent preferences.<\/strong>
\n\u00b7 Technology planned for application in context-aware AI and next-generation agentbased recommendation systems.<\/strong><\/p>\n

\"\"<\/p>\n

SK Telecom (NYSE:SKM, hereinafter referred to as \u201cSKT\u201d) announced today that its new AI recommendation model, which more accurately identifies customers\u2019 true preferences, was presented at a leading AI conference, the International Conference on Learning Representations (ICLR) 2026.<\/p>\n

ICLR, along with the Conference on Neural Information Processing Systems (NeurIPS) and the International Conference on Machine Learning (ICML), is regarded as one of the world\u2019s top three AI conferences. Technologies presented at these events see industrial application within a few years or become benchmarks for academic research and product development, making them key indicators of the future direction of AI.<\/p>\n

The research paper, titled \u201cMore Than What Was Chosen: LLM-Based Explainable Recommendation Beyond Noisy User Preferences,\u201d centers on SKT\u2019s proprietary ConflictAware Direct Preference Optimization (C-APO) technology. The technology filters out conflicting preference signals, enabling more accurate recommendations of products that customers genuinely desire.<\/p>\n

To achieve this, SKT introduced the concept of \u201ccoherent preference\u201d \u2014 reflecting recurring patterns and contextual coherence across a customer\u2019s overall behavior. Through this approach, the model distinguishes coherent preferences from revealed preferences derived from single actions, such as accidental clicks or one-time, unintentional choices.<\/p>\n

This approach increases the learning weight when the two preference signals align, reinforcing the recommendation. When the signals conflict, the model treats the discrepancy as noise and lowers the learning weight, enabling it to identify the most appropriate recommendations.<\/p>\n

Conventional AI recommendation systems have struggled to interpret conflicting preference signals \u2014 for example, a customer who typically watches romance films may receive only action recommendations after watching even a single action movie based on a friend\u2019s suggestion. SKT\u2019s new recommendation model, however, filters out such inconsistencies to identify the customer\u2019s true preferences.<\/p>\n

SKT plans to apply this research to future agent-based customer recommendation systems that integrate AI for interpreting usage context, generating recommendation candidates, and refining results through an orchestration layer. The company aims to deliver more advanced and personalized recommendations tailored to each customer\u2019s needs and behavior.<\/p>\n

\u201cThe technology lays the foundation for delivering personalized recommendations by better understanding customers\u2019 true preferences,\u201d said Seok Ji-hwan, Head of AI\/DT Data Office at SK Telecom. \u201cThe presentation at ICLR 2026 marks a milestone in SKT\u2019s AI transformation to enhance customer experience, and we will continue to advance AI technologies that customers can benefit from.\u201d<\/p>\n

\"\"<\/p>\n","protected":false},"excerpt":{"rendered":"

\u00b7 Paper on AI recommendation model identifying customers\u2019 true preferences presented at ICLR 2026. \u00b7 Proprietary Conflict-Aware Direct Preference Optimization (C-APO) technology improves recommendation performance by distinguishing customers\u2019 revealed and coherent preferences. \u00b7 Technology planned for application in context-aware AI<\/p>\n","protected":false},"author":3,"featured_media":3016,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[4],"tags":[24,540,539,541],"class_list":["post-3015","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-press-release","tag-ai","tag-c-apo","tag-iclr2026","tag-neurips"],"acf":[],"_links":{"self":[{"href":"https:\/\/news.sktelecom.com\/en\/wp-json\/wp\/v2\/posts\/3015","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/news.sktelecom.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/news.sktelecom.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/news.sktelecom.com\/en\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/news.sktelecom.com\/en\/wp-json\/wp\/v2\/comments?post=3015"}],"version-history":[{"count":2,"href":"https:\/\/news.sktelecom.com\/en\/wp-json\/wp\/v2\/posts\/3015\/revisions"}],"predecessor-version":[{"id":3021,"href":"https:\/\/news.sktelecom.com\/en\/wp-json\/wp\/v2\/posts\/3015\/revisions\/3021"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/news.sktelecom.com\/en\/wp-json\/wp\/v2\/media\/3016"}],"wp:attachment":[{"href":"https:\/\/news.sktelecom.com\/en\/wp-json\/wp\/v2\/media?parent=3015"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/news.sktelecom.com\/en\/wp-json\/wp\/v2\/categories?post=3015"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/news.sktelecom.com\/en\/wp-json\/wp\/v2\/tags?post=3015"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}