Author

Xiaohongshu Tech REDtech

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

119

Articles

Likes

409

Views

Comments

Latest from Xiaohongshu Tech REDtech

100 recent articles max

Xiaohongshu Tech REDtech

Jan 23, 2024 · Artificial Intelligence

Controllable Mind Visual Diffusion Model (CMVDM) for Reconstructing Visual Stimuli from fMRI Signals

The Controllable Mind Visual Diffusion Model (CMVDM) decodes fMRI signals into semantic vectors and silhouette maps, feeds them into a latent diffusion framework with a ControlNet‑style encoder, and reconstructs high‑fidelity images that surpass existing baselines in both structural similarity and semantic accuracy across multiple brain‑imaging datasets.

AIbrain decodingdiffusion model

0 likes · 13 min read

Controllable Mind Visual Diffusion Model (CMVDM) for Reconstructing Visual Stimuli from fMRI Signals

Xiaohongshu Tech REDtech

Jan 20, 2024 · Artificial Intelligence

Decoding Xiaohongshu’s Recommendation System: How Ordinary Users Gain Visibility

Xiaohongshu’s recommendation system uses large‑scale multimodal embeddings, dual‑tower and graph models, and diversity techniques like DPP and SSD to quickly surface high‑quality user‑generated content, enabling ordinary users to gain visibility while balancing personalization, exploration, and efficient LLM‑augmented pipelines.

Xiaohongshucold startcontent diversity

0 likes · 15 min read

Decoding Xiaohongshu’s Recommendation System: How Ordinary Users Gain Visibility

Xiaohongshu Tech REDtech

Jan 18, 2024 · R&D Management

Beyond Coding: Engineer Self‑Iteration and Growing with the Organization

In a fast‑changing workplace, engineers must pursue self‑success, cooperation and empowerment by developing small and big T‑shaped skills, mastering clear goal‑setting, meeting structures, and the WWH perspective, while leveraging AIGC as a tool, to continuously iterate toward leadership and meaningful career growth.

LeadershipT-shaped talentcareer development

0 likes · 14 min read

Beyond Coding: Engineer Self‑Iteration and Growing with the Organization

Xiaohongshu Tech REDtech

Jan 12, 2024 · Artificial Intelligence

Negative Sample Assisted Distillation for Large Language Models

The AAAI‑2024 paper introduces a Negative Sample Assisted Distillation framework—comprising Negative Assistance Training, Negative Calibration Enhancement, and Adaptive Self‑Consistency—that leverages both correct and incorrect reasoning examples to train a compact LLaMA‑7B student, achieving up to 75.75 % accuracy gains over fine‑tuning on MATH and improving out‑of‑domain benchmarks.

Knowledge DistillationLLMchain-of-thought

0 likes · 13 min read

Negative Sample Assisted Distillation for Large Language Models

Xiaohongshu Tech REDtech

Dec 14, 2023 · Cloud Native

Evolution of Xiaohongshu Metrics System: Cloud‑Native Observability, High Availability, and Performance Optimizations

Xiaohongshu’s observability team rebuilt its Prometheus‑based metrics platform using vmagent, dual‑active HA clusters, query push‑down, high‑cardinality governance and multi‑cloud active‑active design, delivering ten‑fold collection speed, up to 70× query capacity, massive CPU‑memory‑storage savings and fully automated scaling.

MetricsTime-seriesVictoriaMetrics

0 likes · 35 min read

Evolution of Xiaohongshu Metrics System: Cloud‑Native Observability, High Availability, and Performance Optimizations

Xiaohongshu Tech REDtech

Dec 13, 2023 · Information Security

XiaoHongShu’s Zero Trust SASE Office Security Solution: Architecture, Challenges, and Implementation

XiaoHongShu’s award‑winning Zero‑Trust SASE Office Security Solution integrates all‑in‑one DLP, anti‑virus, identity‑bound access and distributed POPs to replace fragmented agents, delivering real‑time risk control, data‑non‑landing protection, multi‑level disaster recovery, and achieving 100% device coverage, an 80% reduction in data leakage and a 70% internal NPS after one year.

Data SecurityOffice SecuritySASE

0 likes · 14 min read

XiaoHongShu’s Zero Trust SASE Office Security Solution: Architecture, Challenges, and Implementation

Xiaohongshu Tech REDtech

Dec 8, 2023 · Artificial Intelligence

Top 100 Global Software Case Study Summit: AGI Era Innovations

The Top 100 Global Software Case Study Summit in Beijing showcased 100 cutting‑edge case studies on technical, product, and organizational innovation, highlighting AGI‑era breakthroughs such as multimodal AI and AIGC, with insights from leaders like Xiaohongshu’s CTO Feng Di, Tong Dejia, and Xu Xiangkai, and offered free tickets through Small Red Book to help enterprises navigate AI‑driven transformation.

AGIAIInnovation

0 likes · 6 min read

Top 100 Global Software Case Study Summit: AGI Era Innovations

Xiaohongshu Tech REDtech

Dec 4, 2023 · Artificial Intelligence

Xiaohongshu Search Engine Innovations Presented at SIGIR-AP 2023

At SIGIR‑AP 2023 in Beijing, Xiaohongshu’s technical team unveiled four key innovations—advanced user‑intent analysis via multi‑stage LLM pre‑training, multimodal vector retrieval, generative inverted‑index enhancements, and a three‑stage relevance‑ranking pipeline with knowledge distillation—to tackle high multi‑intent, long‑tail, and multimodal search challenges for its 260 million‑user platform.

SIGIR-APVector RetrievalXiaohongshu

0 likes · 13 min read

Xiaohongshu Search Engine Innovations Presented at SIGIR-AP 2023

Xiaohongshu Tech REDtech

Nov 27, 2023 · Cloud Native

Mixed-Workload Scheduling and Resource Utilization Optimization in Xiaohongshu's Cloud-Native Platform

Xiaohongshu’s cloud‑native platform adopted a four‑stage mixed‑workload scheduling strategy—reusing idle nodes, whole‑machine time‑sharing, normal mixed pools, and a unified scheduler (Tusker) that coordinates CPU, GPU and memory across Kubernetes and YARN—boosting average cluster CPU utilization from under 20 % to over 45 % and delivering millions of low‑cost core‑hours while preserving QoS for latency‑sensitive, mid, and batch jobs.

Big DataKubernetesQoS

0 likes · 19 min read

Mixed-Workload Scheduling and Resource Utilization Optimization in Xiaohongshu's Cloud-Native Platform

Xiaohongshu Tech REDtech

Nov 6, 2023 · Artificial Intelligence

Large Models and Recommendation Systems: Challenges, Opportunities, and Future Directions

At CNCC 2023, leading researchers and industry experts convened to examine how large language models can transform recommendation systems, outlining four core challenges—model integration, fluency versus intelligence, hallucination versus deception, and user understanding—while highlighting opportunities such as multimodal content, cold‑start solutions, zero‑shot ranking, instruction‑driven algorithms, and responsible, interactive recommendation pipelines.

AICNCC 2023LLM applications

0 likes · 16 min read

Large Models and Recommendation Systems: Challenges, Opportunities, and Future Directions