Tencent's official tech account. Delivering quality technical content to serve developers.
The article explains why large language models generate hallucinations—due to data errors, training conflicts, and inference uncertainty—and outlines data‑cleaning, model‑level feedback, knowledge augmentation, constraint techniques, and post‑processing methods such as the “Truth‑seeking” algorithm to mitigate the issue.
This article explains what vectors and vector databases are, how they differ from traditional databases, their core technologies, their relationship with large language models, market trends, and details of Tencent Cloud VectorDB’s capabilities, architecture, real‑world applications, and future competitive challenges.
Tencent Robotics X unveiled how its robot dog Max combines pre‑trained AI models with reinforcement learning across three learning stages, enabling it to acquire, store, and apply skills for autonomous decision‑making in complex tasks such as the World Chase Tag competition.
WeChat’s palm‑payment combines unique palm‑texture and vein data with advanced imaging and AI algorithms to verify identity, offering large‑scale deployment, seamless user experience, and strong anti‑spoofing measures while addressing real‑world challenges like dirty or injured hands.
On April 25, Tencent’s Robotics X lab unveiled its self‑developed TRX‑Hand and TRX‑Arm, a three‑finger dexterous hand and a seven‑degree‑of‑freedom arm that combine advanced hybrid actuation, high‑resolution tactile sensors, multi‑modal perception and real‑time planning to perform complex tasks such as cocktail mixing, object catching, and precise insertion.
Tencent’s home‑grown Zixiao AI inference chip, combined with the LightRuntime engine, dramatically reduces latency and cost for real‑time subtitles in Tencent Meeting, handling tens of thousands of concurrent audio streams while meeting sub‑second delay requirements through hardware‑software co‑optimizations and mixed‑precision model tuning.
At GDC, Tencent AI Lab demonstrated an AIGC‑driven pipeline that automatically creates massive, realistic 3D city environments—including road networks, diverse building facades, and interior mappings—in weeks instead of years, dramatically accelerating AAA game development.
Tencent AI Lab deputy Yu Dong was named a 2022 ACM Fellow for his groundbreaking work in speech processing and deep‑learning applications, boasting over 100 patents, multiple best‑paper awards, and technologies now embedded in many of Tencent’s products.
This article examines how the viral mini‑game "Sheep..." overcame its initial 5,000‑QPS bottleneck and scaled to over 100 million daily active users by redesigning its architecture, implementing cloud‑native auto‑scaling, enhancing operational monitoring with CLS, and fortifying security with WAF.
Tencent Cloud’s self‑developed HARP protocol delivers ultra‑high‑performance, highly available and scalable data‑center networking, supporting over 10,000 nodes, microsecond‑level fault recovery, and up to 200 Gbps throughput, making it ideal for AI training, storage and big‑data workloads.
