Xiaohongshu Tech REDtech
Author

Xiaohongshu Tech REDtech

Official account of the Xiaohongshu tech team, sharing tech innovations and problem insights, advancing together.

115
Articles
0
Likes
324
Views
0
Comments
Recent Articles

Latest from Xiaohongshu Tech REDtech

100 recent articles max
Xiaohongshu Tech REDtech
Xiaohongshu Tech REDtech
Aug 17, 2023 · Artificial Intelligence

Human Visual Perception Based Edge‑Cloud Super‑Resolution Framework and RedVQA Quality Assessment at XiaoHongShu

At LiveVideoStackCon 2023, XiaoHongShu unveiled an edge‑cloud super‑resolution framework guided by the human‑perception‑aligned RedVQA model, which jointly optimizes video quality and bandwidth by generating a dedicated SR bitrate tier in the cloud and applying a lightweight SR algorithm on the client, achieving notable QoE gains and narrow‑band HD delivery.

AISuper-Resolutionbandwidth optimization
0 likes · 25 min read
Human Visual Perception Based Edge‑Cloud Super‑Resolution Framework and RedVQA Quality Assessment at XiaoHongShu
Xiaohongshu Tech REDtech
Xiaohongshu Tech REDtech
Jul 26, 2023 · Industry Insights

Human‑Perception‑Based End‑Cloud Super‑Resolution: Cutting Bandwidth, Boosting Quality

The LiveVideoStackCon 2023 session revealed how a human‑perception‑driven end‑cloud super‑resolution framework, AI‑based no‑reference video quality assessment, and rigorous AB‑testing methods can dramatically reduce video bandwidth while enhancing visual quality, illustrating the broader challenges and opportunities in modern audio‑video systems.

AB testingAI assessmentSuper-Resolution
0 likes · 13 min read
Human‑Perception‑Based End‑Cloud Super‑Resolution: Cutting Bandwidth, Boosting Quality
Xiaohongshu Tech REDtech
Xiaohongshu Tech REDtech
Jun 20, 2023 · Artificial Intelligence

Open-Vocabulary Object Attribute Recognition with OvarNet: A Unified Framework for Detection and Attribute Classification

At CVPR 2023 the Xiaohongshu team presented OvarNet, a unified one‑stage Faster‑RCNN model built on CLIP that uses prompt learning and knowledge distillation to jointly detect objects and recognize open‑vocabulary attributes, achieving state‑of‑the‑art results on VAW, MS‑COCO, LSA and OVAD datasets.

Multimodal Learningattribute recognitioncomputer vision
0 likes · 12 min read
Open-Vocabulary Object Attribute Recognition with OvarNet: A Unified Framework for Detection and Attribute Classification
Xiaohongshu Tech REDtech
Xiaohongshu Tech REDtech
May 15, 2023 · Artificial Intelligence

GPU-Accelerated Inference Optimization for Large-Scale Machine Learning at Xiaohongshu

Xiaohongshu transformed its recommendation, advertising, and search inference pipeline by migrating to GPU‑centric hardware, deploying a custom TensorFlow‑Core Lambda service, and applying system‑level, virtualization, and compute‑level optimizations—including NUMA binding, kernel fusion, dynamic scaling, and FP16 quantization—achieving roughly 30× compute capacity growth, over 10% user‑metric gains, and more than 50% cluster‑resource savings.

GPU OptimizationHardware AccelerationMachine Learning Inference
0 likes · 20 min read
GPU-Accelerated Inference Optimization for Large-Scale Machine Learning at Xiaohongshu
Xiaohongshu Tech REDtech
Xiaohongshu Tech REDtech
Apr 25, 2023 · Databases

REDtao: A High-Performance Graph Storage System for Social Graph Data

REDtao is a high‑performance graph storage system built for Xiaohongshu that extends Facebook’s Tao architecture with a three‑layer hash structure, decoupled caching, leader‑follower distribution and cross‑cloud availability, delivering over 90% cache hits, 70% MySQL CPU reduction, 150 M QPS on a 16‑core server and seamless migration despite a 250% surge in daily‑active‑user traffic.

Cachingcloud-nativedistributed system
0 likes · 17 min read
REDtao: A High-Performance Graph Storage System for Social Graph Data
Xiaohongshu Tech REDtech
Xiaohongshu Tech REDtech
Apr 3, 2023 · Industry Insights

What Drives Intelligent Recommendation and Search? Key Takeaways from Xiaohongshu’s CCF C³ Event

The CCF C³ event at Xiaohongshu gathered leading researchers and industry experts to dissect the latest advances, challenges, and future opportunities in intelligent recommendation and search, including multimodal content handling, decentralized distribution, cold‑start solutions, and the impact of large language models.

AIRecommendation SystemsSearch
0 likes · 11 min read
What Drives Intelligent Recommendation and Search? Key Takeaways from Xiaohongshu’s CCF C³ Event
Xiaohongshu Tech REDtech
Xiaohongshu Tech REDtech
Mar 21, 2023 · Artificial Intelligence

From Daily to Minute-Level Updates: Real-Time Recommendation System Enhancements at Xiaohongshu

Xiaohongshu transformed its recommendation pipeline from daily to minute‑level updates by redesigning recall, ranking and feature‑joining components, deploying a base‑plus‑incremental training scheme, migrating Spark to Flink, rewriting services in C++, and optimizing RocksDB, which yielded over 10% longer dwell time, 15% more interactions and roughly 50% higher new‑note efficiency.

Model ServingReal-time Traininglarge‑scale systems
0 likes · 20 min read
From Daily to Minute-Level Updates: Real-Time Recommendation System Enhancements at Xiaohongshu
Xiaohongshu Tech REDtech
Xiaohongshu Tech REDtech
Mar 16, 2023 · Industry Insights

How Xiaohongshu Secured a Top Spot in Shanghai’s “Panshi” Cybersecurity Drill

On March 10, Shanghai’s telecom and internet regulators held a summary conference for the 2022 “Panshi Action” cyber‑attack‑defense drill, where Xiaohongshu ranked third among 123 defending teams, earning awards for excellence and progress while showcasing its incident‑response and traceability capabilities.

CybersecurityShanghaiXiaohongshu
0 likes · 4 min read
How Xiaohongshu Secured a Top Spot in Shanghai’s “Panshi” Cybersecurity Drill
Xiaohongshu Tech REDtech
Xiaohongshu Tech REDtech
Feb 16, 2023 · Artificial Intelligence

Intelligent Creative Generation and Optimization for Xiaohongshu Advertising

Xiaohongshu’s end‑to‑end intelligent creative platform extracts high‑quality images, generates diverse titles with RED‑pretrained GPT‑2/T5 models, and selects the best ads using a UCB‑based multi‑armed bandit that balances CTR uplift, revenue and user‑experience, while employing position‑corrected metrics and a scalable dual‑tower DNN to boost long‑tail performance and overall revenue.

AIAdvertisingNLP
0 likes · 18 min read
Intelligent Creative Generation and Optimization for Xiaohongshu Advertising
Xiaohongshu Tech REDtech
Xiaohongshu Tech REDtech
Feb 10, 2023 · Artificial Intelligence

Expert Insights on ChatGPT: Technical Challenges, Applications, and Future Directions

In a REDtech live interview, NLP professor Li Lei and Xiaohongshu engineers examined ChatGPT’s strengths—long, topic‑focused replies and few‑shot learning—and its challenges such as hallucinations, safety, lack of real‑time data, model compression, and multimodal AIGC, outlining how the technology could reshape content creation, customer service, and search while requiring careful risk management.

AIAI safetyChatGPT
0 likes · 20 min read
Expert Insights on ChatGPT: Technical Challenges, Applications, and Future Directions