This article explains how Douyin leverages large language models to build an end‑to‑end user‑experience pipeline that detects signals, understands feedback, attributes issues, and automates governance, turning reactive fixes into proactive, data‑driven product improvements.
This article presents JD.com’s multimodal recommendation system architecture, covering content understanding, multimodal ranking and recall models, practical deployment pipelines, and future research directions such as large‑model integration and supply‑side generation, all illustrated with detailed diagrams and Q&A.
This article presents JD's multimodal content‑understanding framework, detailing its five‑M business characteristics, the architecture of multimodal recall and ranking models, the GMF and MIN modules for semantic alignment and personalization, and future research directions involving large language models and end‑to‑end multimodal encoding.
This article details Baidu Search's engineering practice for trillion‑scale content understanding, covering cost and efficiency challenges, model‑service framework, batch‑compute platform, resource‑scheduling system, HTAP storage design, and concrete optimization techniques such as multi‑process Python serving, dynamic batching, and two‑stage scheduling.
The article presents Tencent Advertising's comprehensive multimedia AI platform—including content understanding, intelligent creation, automated review, fingerprinting, and a large multimodal model—detailing their architectures, key techniques, and the significant performance and efficiency gains achieved across the ad ecosystem.
Liu Wei outlines Tencent’s Advertising Multimedia AI ecosystem on the Taiji platform, describing a five‑platform matrix—Jue for content understanding, Qiankun for automated video creation, Shenzhen for AI‑driven review, Tianyin for hierarchical fingerprinting, and Hunyuan as a multimodal large model—featuring innovations such as massive multimodal pre‑training, logo retrieval, QA‑style attribute extraction, spatiotemporal video analysis, advanced auto‑judgment, and high‑performance hashing that achieve top cross‑modal retrieval results.
This article presents JD.com's end‑to‑end intelligent creative system, covering the background of content‑driven e‑commerce, a multi‑stage content understanding pipeline, AI‑powered video, image and copy generation, multimodal creative selection and distribution, and real‑world business impact.
This talk presents JD's intelligent creative ecosystem for advertising, detailing the construction of a content understanding system, AI-driven content generation (including OCR, image tagging, video summarization, and copy generation), and multimodal creative selection and distribution, highlighting challenges, solutions, and business impact.
The article examines cold‑start challenges for rich‑media recommendations, outlines detection via calibration and lifecycle monitoring, and proposes two remedies—the multi‑stage “rise channel” for promoting fresh content and cross‑modal understanding using CLIP, CB2CF and dual‑tower models—demonstrating NetEase Cloud Music’s 25% distribution boost, over 20% CTR rise, and 40% review‑work reduction.
This article presents QQ Music’s end‑to‑end solution for data‑driven content understanding, value evaluation, and fine‑grained operation, detailing offline and real‑time pipelines, neural‑network models, a content middle‑platform, parameter services, and a precise delivery system that boost user engagement while preserving experience.
This article explains how large‑scale UGC data is used to build a multi‑type interest point graph, describes the mining, hierarchical and associative relationship extraction methods, and demonstrates how the graph improves content understanding and recommendation accuracy while mitigating filter‑bubble effects.
The article explains how Tencent addresses the limitations of traditional content understanding methods in personalized recommendation by introducing an interest‑graph framework that combines classification, concept, entity, and event layers, and details the associated mining, matching, and online evaluation techniques.
This article explains how Tubi leverages machine learning—particularly PyTorch, Databricks, and cloud services—to improve content understanding, advertising technology, and recommendation systems within its advertising‑based video‑on‑demand platform, outlining the three AVOD pillars, technical stack, and future research directions.
This article details the technical architecture behind WeChat's 'Kan Kan' content understanding platform, covering text and multimedia analysis, tag extraction, entity recognition, knowledge graph construction, and how these components enhance recommendation recall, ranking, and user engagement across the ecosystem.
This article explains how Weibo's advertising platform leverages content understanding—covering system architecture, problems caused by insufficient comprehension, the construction of NLP and vision capabilities, content‑based ad strategies, and a celebrity‑brand knowledge graph—to improve ad relevance and ROI.
Leveraging AI to analyze visual, audio, and textual cues, this guide details how video ads—pre‑, mid‑, post‑roll, overlays, and product placements—are recognized, generated, and optimized for usefulness, naturalness, and prominence through multimodal pipelines, algorithmic representations, and a scalable delivery architecture.
The article explains how Tencent tackles content understanding in feed recommendation by evolving from traditional classification, keyword, and entity methods to a multi‑layer interest graph that captures concepts and events, addressing the need for full context, reasoning about user intent, and improving online performance.
This article explains how Tencent tackles content understanding for personalized feed recommendation by combining traditional classification, keyword, and entity methods with deep learning embeddings, introducing an interest graph composed of taxonomy, concept, entity, and event layers to capture full context and infer user consumption intent.
This article presents a comprehensive overview of the challenges, requirements, and technical solutions for multilingual content understanding in UC's international information‑flow recommendation system, covering structured signal construction, low‑resource NLP techniques, transfer learning, quality modeling, and image‑based signal integration.
Youku’s Algorithm Center has built a multimodal AI pipeline that jointly processes visual, audio, and textual signals to enhance video search, recommendation, and digital asset management, overcoming traditional keyword limits, improving relevance and cold‑start issues, while tackling fusion, cost, and interpretability challenges.
This article details Hulu's AI-driven techniques for fine-grained video segmentation, end‑cap detection, subtitle detection and language recognition, background‑music classification, automated processing pipelines, tag generation, and cover‑image regeneration, illustrating how these methods improve user experience and operational efficiency.
On August 25, iQIYI’s Chengdu R&D Center hosted its second Technical Salon, featuring talks on AI-driven content understanding for short‑video feeds, speech synthesis and editing, industry‑standard speech recognition, semantic search ranking, anti‑spam UGC text analysis, and concluding with recruitment invites and a preview of the upcoming Shanghai salon.
The article reviews Hulu’s 2018 iQIYI keynote on AI video applications, detailing how AI drives personalized recommendations, content analysis through computer vision and NLP, ad targeting across visual, linguistic, and semantic layers, and outlines the platform’s machine‑learning architecture and future directions.
The article announces a PhD workshop, outlines Hulu’s research center and its six AI‑focused teams—personalization, data science, video codec, content understanding, intelligent search, and ad intelligence—while highlighting key projects and inviting PhD candidates to apply.
