Official tech channel of 360, building the most professional technology aggregation platform for the brand.
This article examines ChatGPT's limitations in real‑time information and answer accuracy, then proposes a retrieval‑augmented method that combines up‑to‑date document search with large language models to deliver more reliable and current responses across various scenarios.
This article explains why performance stress testing is essential, describes core metrics and the wrk tool, outlines a step‑by‑step testing workflow, analyzes bottlenecks, and presents concrete optimization measures that doubled the single‑machine QPS from 300 to 600 for a Nuxt‑based web service.
This article explains the concepts of characters, character sets, and character encoding, describes how computers store and render text using methods like ASCII, GB2312, Unicode, and UTF‑8/16/32, and discusses why garbled text occurs across different languages and systems.
The CSIG‑hosted "Enterprise Visit – Into Qihoo 360" event on June 29, 2023 gathered over a thousand participants to explore multimodal and cross‑modal learning in the large‑model era, featuring keynote speeches from leading university researchers and Qihoo 360 AI experts, a tour of the company's facilities, and discussions on future AI research directions.
This article explains the four primary browser rendering approaches—DOM (HTML+CSS), SVG, Canvas, and WebGL/WebGPU—detailing their principles, advantages, limitations, and practical examples to help developers understand the underlying mechanics of web‑based visualizations.
The article reviews why visual ability is essential for artificial general intelligence, compares native multimodal and expert‑stitching integration approaches, details the architectures of models such as KOSMOS‑1, PALM‑E, Flamingo, BLIP‑2, LLAVA, miniGPT‑4, and introduces the SEEChat project that fuses CLIP vision encoders with chatGLM6B via a projection layer, presenting its training pipeline, experimental results, and future directions.
This article explains the principles of Interactive Application Security Testing (IAST), details the JavaAgent-based agent instrumentation approach, and provides a step‑by‑step guide to building a Java IAST agent using Javassist, custom class loaders, ThreadLocal tracing, and Maven packaging.
This article provides a comprehensive overview of WebRTC’s MediaStream and RTCPeerConnection APIs, covering concepts such as sources, sinks, tracks, device enumeration, media constraints, resolution and bitrate settings, compatibility issues, screen sharing, content hints, and step‑by‑step connection establishment for real‑time communication in browsers.
The article outlines how a unified, big‑data‑based user profiling platform addresses traditional data silos, high costs, and limited functionality by standardizing tags, integrating Spark and RHadoop processing, and enabling a closed‑loop marketing workflow that improves accuracy and operational efficiency.
This article analyzes the challenges of distributed transactions in a micro‑service order system, compares XA, TCC, and SAGA patterns, and details how the open‑source DTM framework is applied to achieve reliable order creation while handling crashes, rollbacks, idempotency, and network anomalies.
