#knowledge accumulation
553 related articles
On-Policy Distillation Explained: Prin…
·3 min
On-Policy Distillation Explained: Principles, Advantages, and Applications — A Comprehensive Guide
A deep dive into On-Policy Distillation: core principles, key differences from Off-Policy methods, and applications in model compression, reasoning transfer, RLHF alignment, and self-improvement.
Read more →
·2 min
Gemini 3.5 Live Translate: A Deep Dive into Google's Real-Time Translation Model
Google releases Gemini 3.5 Live Translate, a real-time audio translation model supporting multilingual low-latency speech translation. A deep dive into its tech, use cases, and industry impact.
Read more →
·2 min
AlphaFold Key Figure John Jumper Departs: A Nine-Year Partnership with Demis Hassabis Comes to an End
DeepMind's Demis Hassabis bids farewell to AlphaFold lead John Jumper after 9 years. A look at their journey from protein folding breakthroughs to the Nobel Prize in Chemistry.
Read more →
·4 min
Sakana AI Enters the Defense Sector: How Generative AI Is Reshaping National Defense Software Development
Sakana AI launches Applied Team to bring generative AI to defense C2 systems and disinformation countermeasures. Deep dive into DDIL challenges, human-AI collaboration principles, and team culture.
Read more →
·3 min
Sakana AI Enters Defense Intelligence: Partnering with Japanese Think Tank DEEP DIVE to Advance AI-Powered Intelligence Analysis
Sakana AI partners with Japanese think tank DEEP DIVE to apply AI to defense intelligence analysis, combining OSINT data with AI capabilities to overcome human analysis bottlenecks.
Read more →
·3 min
NVIDIA ACE SDK: On-Device AI Inference for Intelligent Game NPC Companions
Deep dive into NVIDIA ACE Game Agent SDK's integration with Unreal Engine 5, exploring how on-device AI inference enables low-latency, privacy-safe intelligent NPC dialogue and behavior.
Read more →
·2 min
AI Generates 7 Cocos Dissolve Shaders at Once — Zero Errors, Ready to Use
Generate 7 Cocos grayscale dissolve Shader versions at once with AI templates — zero compilation errors, ready to use. Covers dissolve principles and AI-assisted Shader workflows.
Read more →
·4 min
AI Agent Development: A Complete Six-Week Systematic Learning Roadmap
A systematic six-week learning roadmap for AI Agent development covering core architecture, ReAct paradigm, multi-agent collaboration, RAG integration, deployment, and hands-on projects.
Read more →
·3 min
The Complete Skill Map for AI Application Engineers: A Progressive Roadmap from Fundamentals to Production
A systematic breakdown of the complete skill structure for AI application engineers, covering Python & deep learning fundamentals, small model engineering, LLM fine-tuning, Agent development, and enterprise projects.
Read more →
·3 min
OpenAI o3 Diagnoses Rare Childhood Diseases: A Deep Dive into the NEJM AI Study
OpenAI and Boston Children's Hospital published research in NEJM AI showing how the o3 Deep Research model helps clinicians diagnose previously unresolved rare childhood diseases.
Read more →
GPT-5.4 Powers a Drug Chemistry Projec…
·3 min
GPT-5.4 Powers a Drug Chemistry Project: A Complete Closed Loop from Literature Review to Experimental Validation
GPT-5.4 teams with Maria AI and automated labs to complete a full drug chemistry closed loop—from literature review and hypothesis generation to experimental validation.
Read more →
·3 min
Anthropic Skills Mechanism Explained: An Open Standard for Giving AI Installable Skill Packs
Deep dive into Anthropic Skills (nearly 150K GitHub stars): skill.md structure, progressive loading, ready-to-use skills ecosystem, and the cross-tool open standard for AI skill packs.
Read more →
·3 min
In-Depth Guide to the Codex Chinese Manual: A Complete Walkthrough from Beginner to Advanced
In-depth breakdown of ByteDance's 198-page Codex Chinese manual covering installation, Commands, MCP workflows, Skills templates, and multi-Agent collaboration.
Read more →
·3 min
Karpathy's Advanced Claude Code Methodology: Building a Self-Evolving AI Work Environment
Karpathy's Claude Code methodology: build a self-evolving AI environment using CLAUDE.md, knowledge bases, Skills, and Hook guardrails for compounding efficiency.
Read more →
·4 min
DeepSeek Forms Harness Team: AI Coding Competition Enters the Second Half
DeepSeek forms a dedicated Harness team to rival Claude Code. Analysis of the four-layer architecture, three core advantages, and 40x cost edge driving AI competition from model wars to engineering deployment.
Read more →
·2 min
Self-Driving Labs: In AI Materials Science, the Moat Is the Lab, Not the Model
Self-driving labs integrate AI decision-making with automated experiments for closed-loop materials R&D acceleration. Learn why the real moat in AI materials science is the lab, not the model.
Read more →
·4 min
Claude Code in Practice: Complex Payment Feature Built in 4 Hours for $60
A detailed walkthrough of Claude Code + Opus 4.7 in action: from CCSwitch setup to model configuration, completing a complex payment feature in 4 hours for $60 with zero hand-written code.
Read more →
·4 min
UniApp Beginner Tutorial: Build Multi-Platform Apps with One Codebase
In-depth analysis of a UniApp zero-to-hero APP development course covering HTML, CSS, JavaScript, and UniApp, with project-driven teaching and learning path recommendations.
Read more →
·3 min
Enterprise-Grade RAG Full-Pipeline Optimization: Multi-Turn Conversations, Retrieval Tuning, and Production Engineering in Practice
A deep dive into full-pipeline optimization for enterprise RAG systems, covering multi-turn query rewriting, retrieval tuning, and quality evaluation to take RAG from demo to production.
Read more →
·2 min
Extract Skills in One Click During a Claude Code Session: Turn Inspiration into Reusable Skills Instantly
Learn how to extract reusable Skill files in Claude Code without leaving your session. One command preserves full context, turning 20 minutes of work into seconds.
Read more →