Author

AI2ML AI to Machine Learning

Original articles on artificial intelligence and machine learning, deep optimization. Less is more, life is simple! Shi Chunqi

Articles

Likes

288

Views

Comments

Latest from AI2ML AI to Machine Learning

53 recent articles

AI2ML AI to Machine Learning

Oct 24, 2025 · Artificial Intelligence

Beyond RAG: Three Emerging Knowledge‑Engineering Strategies (ICL, Online Learning, SLM)

The article outlines three post‑RAG knowledge‑engineering approaches—In‑Context Learning with dynamic few‑shot selection, Online Learning encompassing Meta‑Learning and Lifelong Learning to quickly adapt to new tasks, and the Small Language Model path that combines fine‑tuned task‑specific experts with LLM‑SLM collaboration for efficient, privacy‑preserving inference.

In-Context LearningLLMLifelong Learning

0 likes · 4 min read

Beyond RAG: Three Emerging Knowledge‑Engineering Strategies (ICL, Online Learning, SLM)

AI2ML AI to Machine Learning

Oct 23, 2025 · Artificial Intelligence

Why Visually‑Rich Document Understanding Looks Like High‑End Docs: A Static Multimodal Overview

The article surveys the evolution of Visually‑Rich Document Understanding (VRDU), highlighting pioneering Chinese OCR research, the LayoutLM family, recent multimodal model breakthroughs, open‑source toolkits, and practical recommendations for handling diverse document types and tasks.

DeepSeek-OCRLayoutLMMultimodal OCR

0 likes · 11 min read

Why Visually‑Rich Document Understanding Looks Like High‑End Docs: A Static Multimodal Overview

AI2ML AI to Machine Learning

Oct 20, 2025 · Artificial Intelligence

nanochat Source Code Deep Dive: Data Prep, Model Design, Training & Evaluation

This article revisits nanochat's core components, detailing the preparation of diverse training datasets, the scaling calculations for tokens and parameters, the model's MQA and KV‑cache design, the full training pipeline with gradient accumulation and mixed‑precision, cost breakdown, inference optimizations, evaluation tasks, and identified limitations with suggested improvements.

KV CacheLLMMQA

0 likes · 9 min read

nanochat Source Code Deep Dive: Data Prep, Model Design, Training & Evaluation

AI2ML AI to Machine Learning

Oct 19, 2025 · Artificial Intelligence

Deep Dive into nanochat: Source Code, Model Size Calculations, and Optimization Techniques

This article provides a thorough analysis of nanochat’s source code, detailing transformer component differences, precise parameter‑size formulas, FlashNorm and ReLU² innovations, scaling‑law insights, memory‑usage estimations, and the distributed optimizer and training pipelines used to build the model.

Distributed TrainingLLMOptimizer

0 likes · 20 min read

Deep Dive into nanochat: Source Code, Model Size Calculations, and Optimization Techniques

AI2ML AI to Machine Learning

Oct 15, 2025 · Artificial Intelligence

NanoChat Source Code Deep Dive: Karpathy’s Full‑Stack LLM Pipeline Explained

This article dissects NanoChat’s end‑to‑end LLM pipeline—from a lightweight 561M‑parameter transformer and custom Rust BPE tokenizer to Chinchilla‑scaled training, multi‑task fine‑tuning, optional RL on GSM8K, KV‑cache inference optimizations, and benchmark results that slightly surpass GPT‑2 Large.

CORE benchmarkChinchilla scalingFastAPI

0 likes · 10 min read

NanoChat Source Code Deep Dive: Karpathy’s Full‑Stack LLM Pipeline Explained

AI2ML AI to Machine Learning

Oct 13, 2025 · Artificial Intelligence

How Large‑and‑Small Language Model Collaboration Is Shaping the Future

The article argues that combining large, high‑capacity models with lightweight, fine‑tuned small models can cut costs, lower latency, enable specialized vertical tasks, and shift development from chasing ever‑bigger models toward optimal system architectures, outlining key techniques such as state‑space models, knowledge distillation, and staged fine‑tuning.

AI ArchitectureModel Collaborationefficiency

0 likes · 3 min read

How Large‑and‑Small Language Model Collaboration Is Shaping the Future

AI2ML AI to Machine Learning

Oct 1, 2025 · Artificial Intelligence

2025 Large Model Engineering Breakthroughs: Cutting Costs, Boosting Performance, and Extending Context

The 2025 open‑source reports reveal major advances in large‑model engineering, including drastic cost cuts such as DeepSeek‑V3 training for $5.57 M, performance gains where Gemma 3 4B matches Gemma 2 27B, memory efficiencies like 85 % KV‑cache reduction, and a suite of new techniques—from loss‑free MoE balancing to multi‑token prediction—that together push context lengths to one million tokens and enable multimodal, aligned, and industry‑specific models.

Attention MechanismsLarge Language ModelsMemory Efficiency

0 likes · 13 min read

2025 Large Model Engineering Breakthroughs: Cutting Costs, Boosting Performance, and Extending Context

AI2ML AI to Machine Learning

Sep 30, 2025 · Artificial Intelligence

Dynamic Multimodal Video Generation: Prioritizing Stability and High Quality

The article surveys the evolution of video generation models—from early GANs and DCGAN to diffusion‑based approaches like Stable Diffusion and DiT—highlighting how stability, high quality, massive compute, and multimodal data pipelines are shaping the current and future paths of dynamic multimodal video generation.

Stable DiffusionTransformerdiffusion models

0 likes · 7 min read

Dynamic Multimodal Video Generation: Prioritizing Stability and High Quality

AI2ML AI to Machine Learning

Sep 28, 2025 · Artificial Intelligence

Core Metrics for Enterprise Large‑Model Engineering

The article outlines the five essential engineering domains—application, model, compute, knowledge, and data—in the era of large models, and details concrete scale, efficiency, service, value, quality, and security metrics that enterprises should track to drive intelligent outcomes.

AI EngineeringData EngineeringModel Performance

0 likes · 7 min read

Core Metrics for Enterprise Large‑Model Engineering

AI2ML AI to Machine Learning

Sep 24, 2025 · Artificial Intelligence

Key Points for Evaluating AI Agents

The article explains how Coze's Compass introduces a flexible evaluation system for AI agents, outlines a four‑dimensional submodule assessment (planning, tool use, self‑reflection, memory), and details specific testing criteria and challenges for web, scientific, dialogue, and programming agents.

AI agentsBenchmarkingCoze

0 likes · 6 min read