Tagged articles
training
53 articles · Page 1 of 1
This comprehensive guide walks through efficient LoRA/QLoRA fine-tuning of 7B and 13B language models on a single 24GB GPU, covering memory budgeting, environment setup, data preparation, training configuration, evaluation strategies, model merging, deployment verification, and production best practices with concrete code examples and benchmarks.
GPUHuggingFaceLLM
0 likes · 78 min read

The article argues that adult mindsets, responsibility, and work standards are deeply ingrained and nearly impossible to change through training; training only teaches skills and processes, while real value lies in unteachable traits, so rigorous upfront selection and early rejection of mismatches save far more time than post-hire coaching.
hiringleadershipmanagement
0 likes · 6 min read

The article provides a detailed technical comparison of SRAM, DRAM and HBM, explains their roles in modern AI GPUs such as the H100, discusses training and inference memory placement, and evaluates emerging ideas like High‑Bandwidth SRAM and High‑Bandwidth Flash, concluding that all three memory types will coexist in future AI systems.
AI hardwareDRAMGPU
0 likes · 12 min read

Unsloth now supports AMD GPUs with custom ROCm‑optimized Triton kernels, delivering up to double the training speed and 70% lower memory usage, enabling over 500 LLMs to be trained on as little as 3 GB VRAM and providing detailed performance benchmarks on MI300X.
AMDGPULLM
0 likes · 5 min read

Data from 脉脉 shows AI positions in 2026 have grown about twelve‑fold year‑over‑year, now accounting for 26.23% of new‑economy jobs with average monthly salaries 26% above peers, while traditional software demand falls 25% and large‑model application roles surge, prompting a training push on RAG and Agent technologies.
AIAgentRAG
0 likes · 3 min read

This survey examines how rubrics—structured, multi‑dimensional evaluation criteria—are defined, constructed, and applied to train and evaluate large language models, especially for open‑ended, high‑risk and agentic tasks, while highlighting current challenges such as reward hacking and bias.
AI safetyAgentReward Modeling
0 likes · 15 min read

The survey from RUC‑Gaoling AI Institute reviews Rubrics for large language models, explaining why they are needed for open‑ended, high‑risk tasks, how they are constructed, and how they can be applied to policy and reward model training as well as multi‑dimensional evaluation across general and domain‑specific scenarios.
AgentLLMRubrics
0 likes · 14 min read

This article clarifies the core concepts of AI agents, distinguishing models from agents, defining scaffolding and harness, and detailing the roles of context engineering, policy, tools, skills, sub‑agents, and training components such as environment, rollout, reward, and trainer.
AI agentLLMPolicy
0 likes · 11 min read

This survey examines how rubrics can decompose the vague notion of a "good answer" for large language models into concrete, multi‑dimensional evaluation criteria, detailing their definition, construction methods, applications in training and evaluation, and the open challenges they present.
AI AlignmentAgentic AIRubrics
0 likes · 13 min read


SuanNiApr 29, 2026 · Artificial Intelligence
Google’s Cloud Next 2026 reveal splits the 8th‑generation TPU into training‑focused Sunfish and inference‑focused Zebrafish, highlighting Ironwood’s record‑breaking performance, a multi‑vendor supply chain, Anthropic’s multi‑gigawatt order, and a broader industry shift toward custom AI chips that promise far higher profit margins than generic GPUs.
AIGoogleTPU
0 likes · 8 min read

Google unveiled its eighth‑generation TPU line at Cloud Next 2026, introducing the training‑focused TPU 8t with a 2.7× performance boost and massive scaling, and the inference‑optimized TPU 8i featuring three‑times more on‑chip SRAM and an 80% performance uplift for agentic AI workloads, while positioning the chips as a complement—not a replacement—to Nvidia's offerings.
AI hardwareAgentic AIGoogle Cloud
0 likes · 9 min read

This article walks readers through the three major training methods—full‑parameter, LoRA, and QLoRA—explaining their memory costs, data requirements, and trade‑offs, then provides a line‑by‑line breakdown of LLaMAFactory configuration files, hyper‑parameter tuning guidelines, and the process for merging LoRA adapters into a deployable model.
LlamaFactoryLoRAModel Merge
0 likes · 27 min read

This guide walks through setting up a Kunlun P800 XPU host, preparing Docker containers, deploying Qwen3‑8B/‑32B/‑VL models with vLLM‑Kunlun, benchmarking performance, and running full‑parameter DPO training using LLaMA‑Factory, providing scripts, configuration files, and troubleshooting tips for AI engineers.
DPOKunlun P800LLaMA-Factory
0 likes · 32 min read

This article surveys the latest research on Unified Multimodal Foundations (UFM), explaining why integrating understanding and generation across text, image, video, and audio is essential for AGI, and detailing modeling paradigms, encoding/decoding strategies, training pipelines, benchmarks, and real‑world applications.
AI researchUnified Modelbenchmark
0 likes · 10 min read

Baidu’s Baige 5.0 AI Computing Platform introduces FP8 mixed‑precision training, MoE‑aware distributed strategies, adaptive parallelism, and a three‑tier KV‑Cache, delivering over 30% training speedup and 50% inference throughput gains while keeping token latency under half a second for large‑scale models.
AIFP8MoE
0 likes · 16 min read

VeOmni, ByteDance’s open‑source unified multimodal training framework, tackles fragmented training pipelines by integrating LoRA fine‑tuning, FSDP, Ulysses, and Expert Parallel, delivering up to 40% higher throughput, up to 55% memory savings, and streamlined one‑click deployment for LLM, VLM, and video models.
AIParallelismframework
0 likes · 14 min read

DeepSeek‑VL2 is a state‑of‑the‑art multimodal model built on a Mixture‑of‑Experts architecture that combines a SigLIP‑L vision encoder with dynamic tiling, a two‑layer VL adaptor, and a DeepSeek‑MoE language model using Multi‑head Latent Attention, trained in three stages on diverse visual‑language and text data, and achieving strong results on benchmarks such as DocVQA and TextVQA, with full implementation and inference code available in PaddleMIX.
DeepSeek-VL2Multimodal ModelPaddleMIX
0 likes · 36 min read

This article details JD.com's challenges and solutions for deploying domestic NPU chips across heterogeneous GPU‑NPU clusters, covering architecture, scheduling, high‑performance training and inference engines, real‑world case studies, and future plans to scale AI workloads securely and efficiently.
AIJD.comNPU
0 likes · 19 min read

This article explains why large language models produce plausible‑looking but false information, traces the problem to the supervised fine‑tuning stage, and outlines mitigation techniques such as knowledge interrogation, RLHF, and tool‑augmented search to reduce hallucinations.
HallucinationLLMRLHF
0 likes · 12 min read


ArchitectFeb 21, 2025 · Artificial Intelligence
This article reviews DeepSeek's recent breakthroughs, including the MLA attention redesign, GRPO alignment algorithm, MoE enhancements, multi‑stage training pipelines (SFT, RL, preference tuning, distillation), and comparative performance against GPT‑4o‑Mini and Llama 3.1, highlighting both strengths and remaining challenges.
DeepSeekModel Evaluationarchitecture
0 likes · 16 min read

This tutorial series provides a step‑by‑step technical guide to training, deploying, and fine‑tuning the MiniDeepSeek V3+R1 large language model, covering model performance, open‑source details, API usage, parameter explanation, multi‑turn chatbot construction, function calling, integration with Open WebUI, GraphRAG, Swarm, and various deployment and optimization techniques.
AIMiniDeepSeekTutorial
0 likes · 4 min read

EasyAnimate V5, an Alibaba Cloud AI video generation framework, expands model size to 7B/12B, introduces multimodal control, token‑length based training, and inpaint‑based image‑to‑video strategies, while providing easy deployment via PAI, DSW, and local ComfyUI integration.
AILoRAMMDiT
0 likes · 11 min read

This article provides a comprehensive, experience‑driven guide to supervised fine‑tuning (SFT) of large language models, covering special tokens, latency considerations, data diversity and production, training frameworks and hyper‑parameters, over‑/under‑fitting diagnostics, and evaluation metrics such as helpfulness, honesty, and harmlessness.
AIData EngineeringLLM
0 likes · 40 min read

The article details JD Advertising's technical challenges and solutions for large‑scale sparse recommendation models, describing GPU‑focused storage, compute and I/O optimizations for both training and low‑latency inference, including distributed pipelines, heterogeneous deployment, batch aggregation, multi‑stream execution, and compiler extensions.
GPU optimizationSparse ModelsTensorFlow
0 likes · 13 min read

The article analyzes why Reinforcement Learning from Human Feedback (RLHF) cannot be replaced by Supervised Fine‑Tuning (SFT), highlighting SFT's lack of negative feedback, its one‑directional attention limitation, and how RLHF's reward models provide crucial safety and performance improvements for large language models.
AI AlignmentRLHFSFT
0 likes · 9 min read

This comprehensive technical report details the development, architecture, training methodology, extensive benchmark evaluations, safety measures, and inference optimizations of Meta's open‑source Llama 3 large language model series, covering models up to 405 billion parameters and supporting multilingual, multimodal, and tool‑use capabilities.
AILlamalarge language model
0 likes · 115 min read

The article traces the evolution of large‑model training and inference infrastructure from the early “black‑iron” era (2019‑2021) through the “golden” boom (2022‑2023) to the emerging “silver” phase (2024‑), highlighting key research breakthroughs, open‑source frameworks, hardware trends, market dynamics, and practical challenges for engineers entering the field.
AI infrastructureindustry trendsinference
0 likes · 22 min read


NewBeeNLPJul 15, 2024 · Artificial Intelligence
This guide compiles open‑source repositories, research papers, and practical tricks for training miniature large‑language models under 1 billion parameters, helping readers learn by reproducing models like nanoGPT, tinyLlama, Phi‑1.5, and more.
LLMnanoGPTopen-source
0 likes · 7 min read


DevOpsJun 24, 2024 · Operations
The article outlines a four‑layer DevOps transformation framework—values, principles, processes, and tools—provides detailed implementation steps, and promotes an official DevOps Engineer certification program with upcoming courses, dates, and career benefits.
Transformationcertificationtraining
0 likes · 5 min read

The article analyzes how generative AI models from GPT‑1 to the upcoming GPT‑5 are driving exponential growth in compute requirements, prompting massive cloud capital expenditures and intense competition among GPU vendors such as NVIDIA, AMD, Google, and emerging domestic chip makers, while also highlighting interconnect innovations and cost‑effective solutions.
AIAcceleratorsCompute
0 likes · 12 min read

Andrej Karpathy’s LLM.c project demonstrates how a compact, pure‑C (and CUDA) codebase of roughly 1000 lines can train a GPT‑2 model, covering data preparation, memory management, layer implementations, compilation, and practical tips for running and testing the model on CPUs and GPUs.
AIC++CUDA
0 likes · 10 min read

The article recounts Yi Tay’s experience founding Reka and building large language models from scratch, highlighting the unpredictable quality of GPU clusters, the challenges of multi‑cluster orchestration, code‑base choices, and how startups must rely on fast, intuition‑driven experimentation to succeed.
GPULLMcluster management
0 likes · 12 min read

This article details Xiaohongshu's end‑to‑end GPU‑based transformation of its recommendation and search models, covering background, model characteristics, training and inference frameworks, system‑level and GPU‑level optimizations, compilation tricks, hardware upgrades, and future directions for large‑scale machine‑learning infrastructure.
GPUModel Servinglarge-scale systems
0 likes · 18 min read

This article analyses the Mixtral‑8x7B Mixture‑of‑Experts LLM, explains its gate‑driven 8‑expert architecture, presents a simplified PyTorch implementation, and reports a series of experiments that probe top‑2 gating during training, individual expert contributions, task‑specific pre‑training, the impact of expert count, and similarity with Mistral‑7B, ultimately offering hypotheses about its training pipeline.
LLMMixtralmixture of experts
0 likes · 14 min read

This article details Xiaohongshu's end‑to‑end GPU‑centric transformation for search‑related machine‑learning models, covering model characteristics, training and inference frameworks, system‑level GPU and CPU optimizations, multi‑card and compilation techniques, and future directions for scaling large sparse and dense models.
GPU optimizationModel ServingXiaohongshu
0 likes · 16 min read

This article compares NVIDIA's high‑end AI GPUs—H100, A100, and the newer L40S—detailing their specifications, performance trade‑offs, pricing, availability, and suitability for training and inference workloads, while highlighting why L40S can be a cost‑effective alternative for many enterprises.
AIGPUH100
0 likes · 10 min read

This article details a company’s year‑long evolution of a one‑day Scrum Master training program, covering its motivation, survey‑driven redesign, scenario creation, theory delivery techniques, domino‑based sandbox exercises, key agile practices like time‑boxing and retrospectives, and the resulting growth of a sizable certified Scrum Master talent pool.
ScrumScrum MasterScrum Workshop
0 likes · 11 min read

In a Stanford eCorner interview, Ilya Sutskever, chief scientist of OpenAI and architect of GPT, reflects on the history of large language models, compares AI to human cognition, debates open‑source versus closed‑source models, discusses regulation, training strategies, and offers guidance for aspiring AI researchers.
AIdeep learninglarge language models
0 likes · 21 min read

The article uses a CEO’s one‑hour flight lesson for ten staff as a metaphor to illustrate why hands‑on, relevant experience is essential for effective industrial control systems (ICS) cybersecurity training, and describes the structure and objectives of the SANS ICS612 course.
Hands‑on ExperienceICSSANS
0 likes · 13 min read

The article analyses the computational resources, hardware costs, and human investment needed to train and serve large language models such as GPT‑3, discusses practical cost calculations, highlights the challenges faced by Chinese AI teams, and argues for sustained, long‑term funding to achieve meaningful breakthroughs.
AI infrastructureChina AIcost estimation
0 likes · 14 min read

This document outlines a detailed Service Level Agreement (SLA) framework covering quality service standards, management processes, testing capabilities, tool support, resource management, measurement systems, risk handling, and continuous improvement to ensure consistent delivery and customer satisfaction across IT operations.
OperationsProcess ImprovementSLA
0 likes · 17 min read


Code DAOJun 7, 2022 · Artificial Intelligence
This article walks through a complete PyTorch implementation of the SRCNN model for image super‑resolution, covering dataset preparation, patch extraction, model architecture, training on a GTX 770 GPU for 2500 epochs, PSNR evaluation, and visual comparisons with bicubic up‑sampling.
PSNRPatchifyPyTorch
0 likes · 22 min read


58UXDApr 29, 2022 · Operations
This article examines how 58 Home Service identified service gaps through user research, built a detailed user‑experience map, created a comprehensive SOP handbook covering image, etiquette, and behavior, and implemented training, assessment, and incentives to dramatically improve customer satisfaction and reduce complaints.
OperationsUser Experiencequality assurance
0 likes · 9 min read

The article outlines a technical exchange hosted by 58.com AI Lab and Tianjin University that discusses high‑efficiency AI computing, resource‑aware scheduling for both online inference and offline training, and methods to mitigate GPU under‑utilization and gray‑interference in distributed deep‑learning platforms.
AIGPU utilizationinference
0 likes · 4 min read


21CTOMay 31, 2021 · Operations
This article recounts how a former programmer became a deep‑learning evangelist, built Huawei’s MindSpore open‑source community, leveraged one‑minute videos, launched a hands‑on training camp and a lightweight tool, and fostered a thriving ecosystem that now serves over 190,000 developers.
AI frameworkCommunity ManagementMindSpore
0 likes · 16 min read

The article explains how to construct an effective DBA operations team by focusing on reusable infrastructure, clear team standards, a structured training system, a comprehensive knowledge base, and a positive team atmosphere, providing practical tools and methods for each aspect.
DBAInfrastructuredatabase operations
0 likes · 4 min read

This article provides a comprehensive overview of federated learning engineering, covering deployment via Docker containers, the design of training and inference frameworks, key services such as communication, training, model management, and registration, and practical considerations for scaling and reliability in production environments.
AIDockerFederated Learning
0 likes · 11 min read


21CTOMar 26, 2020 · R&D Management
The article shares practical insights on managing technical teams in mature companies, covering recruitment strategies, effective interview evaluation, comprehensive training programs, and fostering a positive team culture through democratic decision‑making, balanced freedom, challenging work, innovation encouragement, and mutual support.
CultureRecruitmentleadership
0 likes · 14 min read

This article outlines a systematic approach to onboarding and training new Site Reliability Engineers, covering trust building, readiness assessment, diverse learning methods, structured curricula, on‑call milestones, project‑focused work, reverse‑engineering skills, statistical thinking, and improvisation techniques to develop high‑performing SRE teams.
OperationsSREon-call
0 likes · 17 min read

This guide curates essential books, professional training courses, and major conferences for industrial control systems cybersecurity, offering insights into historical context, technical security practices, and community engagement to help practitioners deepen their knowledge and connect with the field.
ICS securityconferencescybersecurity resources
0 likes · 10 min read

This article provides a humorous yet informative, non‑mathematical guide to deep learning, covering neural network basics, layer addition, training methods, back‑propagation, unsupervised pre‑training, regularization, ResNet shortcuts, GPU computation, framework choices, and practical advice for applying deep learning to industrial data.
AIGPUPu-Learning
0 likes · 26 min read

The announcement details three internal technical training sessions covering XML serialization fundamentals, .NET parallel programming basics, and essential SQL query concepts and optimization, including instructors, topics, locations, target audiences, and scheduled times for each course.
.NETParallel ProgrammingSQL
0 likes · 2 min read

In this interview, senior manager Liang Yonggang of Dongfang Hongtai discusses his company's background, recent eSDK training on UC/IVS/TP, the benefits and shortcomings of the program, suggestions for the eSDK WeChat channel, and valuable software development and project management lessons he has gathered over a decade in the industry.
IVSUCeSDK
0 likes · 7 min read
