AI research, translated and organized.
Chinese paragraph-by-paragraph translations of high-quality AI papers and engineering articles, plus a curated open-source model directory for real project selection.
From paradigm to model selection — how the topics relate
Browse by topic
Six research streams. Every paper is a blueprint you can extract into your project.
Inference Optimization
Parameter offloading, sparsity, quantization, MoE caching — the engineering that makes large models fit where they otherwise wouldn't.
Explore
Self-Improving Agents
From self-play and test-time adaptation to executable subagents and autoresearch ratchets — how agents compound capability over time.
Explore
Agent Patterns
The patterns that frontier labs use when they build real agents — from Anthropic's 5 workflow archetypes to ReAct's foundational loop.
Explore
Training Techniques
Open-source reproducible methods that moved the frontier: GRPO-based reasoning RL (DeepSeek-R1), RLHF replacement via DPO, parameter-efficient adaptation via LoRA.
Explore
Thinking Patterns
Short, dense essays and case studies from frontier practitioners — the mental models actively driving modern engineering decisions. AI-specific paradigms sit alongside general engineering philosophy that applies to AI work.
Explore
Memory Systems
From Ebbinghaus-inspired forgetting curves to LLM-maintained wikis — how agents remember, forget, and compound knowledge over time.
Explore