[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-77684":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":15,"forks30d":15,"starsTrendScore":19,"compositeScore":20,"rankGlobal":9,"rankLanguage":9,"license":21,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":24,"hasPages":24,"topics":25,"createdAt":9,"pushedAt":9,"updatedAt":46,"readmeContent":47,"aiSummary":48,"trendingCount":15,"starSnapshotCount":15,"syncStatus":14,"lastSyncTime":49,"discoverSource":50},77684,"ARIS-in-AI-Offer","wanshuiyin\u002FARIS-in-AI-Offer","wanshuiyin","Bilingual (中文+EN) ML \u002F LLM \u002F diffusion \u002F agent interview cheat sheets for AI 秋招 — generated by ARIS \u002Finterview-cheatsheet, rendered by \u002Frender-html into single-file HTML, reads anywhere — plus a CV→DBLP-fact-checked academic homepage generator and hand-authored long-form blogs 🌱",null,"Python",194,7,1,2,0,4,33,137,23,77.21,"MIT License",false,"main",true,[26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45],"ai-interview","aris","autumn-recruiting","cheatsheet","chinese","claude-code","deep-learning","diffusion","flow-matching","interview-prep","llm","machine-learning","moe","nerf","pytorch","quantization","rlhf","transformer","video-generation","vision-language-model","2026-06-12 04:01:22","\u003Cp align=\"center\">\n  \u003Cimg src=\"assets\u002Faris_logo.svg\" alt=\"ARIS — Auto Research in Sleep\" width=\"640\">\n\u003C\u002Fp>\n\n# ARIS-in-AI-Offer (ARIS in 秋招)\n\n> Hoping to make your **秋招 (qiūzhāo, Chinese AI campus recruiting season)** a little easier 🌱\n>\n> 📖 **中文版 (Chinese version)**: [README_CN.md](README_CN.md)\n\n[![Stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwanshuiyin\u002FAuto-claude-code-research-in-sleep?style=flat&logo=github&logoColor=white&color=gold&label=ARIS%20Stars)](https:\u002F\u002Fgithub.com\u002Fwanshuiyin\u002FAuto-claude-code-research-in-sleep\u002Fstargazers) · [![arXiv](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv-2605.03042-b31b1b?style=flat&logo=arxiv)](https:\u002F\u002Fhuggingface.co\u002Fpapers\u002F2605.03042) · [![HF Daily #1](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHF%20Daily%20Papers-%231-yellow?style=flat)](https:\u002F\u002Fhuggingface.co\u002Fpapers\u002F2605.03042) · [![PaperWeekly](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FFeatured%20on-PaperWeekly-red?style=flat)](https:\u002F\u002Fgithub.com\u002Fwanshuiyin\u002FAuto-claude-code-research-in-sleep) · [![awesome-agent-skills](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FFeatured%20in-awesome--agent--skills-blue?style=flat&logo=github)](https:\u002F\u002Fgithub.com\u002FVoltAgent\u002Fawesome-agent-skills) · [![Project of the Day](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAI%20Digital%20Crew-Project%20of%20the%20Day-orange?style=flat)](https:\u002F\u002Faidigitalcrew.com)\n\n> 🏆 **Built on a battle-tested foundation** — the [**ARIS main repo**](https:\u002F\u002Fgithub.com\u002Fwanshuiyin\u002FAuto-claude-code-research-in-sleep) has ~10k GitHub stars, was HuggingFace Daily Papers #1, won AI Digital Crew Project of the Day, and ships 74+ research skills across 7+ platforms. This isn't a vaporware preview — **every cheat sheet here is the production output** of the same `\u002Finterview-cheatsheet` + `\u002Frender-html` workflow used in academic-research production.\n\nA curated, **bilingual** (中文 + English) collection of ML \u002F LLM \u002F multimodal \u002F diffusion \u002F agent \u002F generative-model interview cheat sheets, auto-generated by the **[ARIS — Auto Research in Sleep](https:\u002F\u002Fgithub.com\u002Fwanshuiyin\u002FAuto-claude-code-research-in-sleep)** `\u002Frender-html` workflow.\n\nEach cheat sheet is a long-form Chinese tutorial with: formula derivations · from-scratch PyTorch code · 25 high-frequency interview questions (L1 essentials · L2 advanced · L3 top-tier lab).\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"assets\u002Fpreview_strip.jpg\" alt=\"ARIS-in-AI-Offer preview — Foundations + Interview Q&A + From-Scratch Code, three columns from a representative cheat sheet\" width=\"100%\">\n\u003C\u002Fp>\n\n> 📖 **Preview** (above): one snapshot per pillar, taken from the [Diffusion Foundations cheat sheet](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fdiffusion_foundations_tutorial.html) — ① **Foundations** (formula derivations + intuition + TL;DR), ② **Interview Q&A** (25 high-frequency questions stratified L1\u002FL2\u002FL3), ③ **From-Scratch Code** (runnable PyTorch, including CFG training + DDIM sampling). Every cheat sheet in this collection follows the same three-pillar structure.\n\n### 📱 HTML reads cleanly everywhere\n\nPhone on the subway, iPad at a café, laptop in the library — same HTML link opens equally well:\n\n- 🧮 **MathJax** renders all LaTeX formulas (**not screenshots** — scalable, copyable, selectable)\n- 💻 **highlight.js** colors all PyTorch code blocks\n- 📐 **Responsive layout** adapts to any window width — no overflow, no blur\n- 📑 **Sticky TOC** for jumping around long documents\n- 💾 **Single-file HTML** — download once, read offline, no backend required\n\n---\n\n## 🌟 What is ARIS — A Quick Pitch\n\n[**ARIS — Auto Research in Sleep**](https:\u002F\u002Fgithub.com\u002Fwanshuiyin\u002FAuto-claude-code-research-in-sleep) is one of the most-watched AI research agent skill platforms of 2025-2026. The `\u002Finterview-cheatsheet` + `\u002Frender-html` skills that produced this repo are 2 out of ARIS's 74+ skills.\n\n[![Stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwanshuiyin\u002FAuto-claude-code-research-in-sleep?style=flat&logo=github&logoColor=white&color=gold&label=Stars)](https:\u002F\u002Fgithub.com\u002Fwanshuiyin\u002FAuto-claude-code-research-in-sleep\u002Fstargazers) · [![arXiv](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv-2605.03042-b31b1b?style=flat&logo=arxiv)](https:\u002F\u002Fhuggingface.co\u002Fpapers\u002F2605.03042) · [![HF Daily #1](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHF%20Daily%20Papers-%231-yellow?style=flat)](https:\u002F\u002Fhuggingface.co\u002Fpapers\u002F2605.03042) · [![PaperWeekly](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FFeatured%20on-PaperWeekly-red?style=flat)](https:\u002F\u002Fgithub.com\u002Fwanshuiyin\u002FAuto-claude-code-research-in-sleep) · [![awesome-agent-skills](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FFeatured%20in-awesome--agent--skills-blue?style=flat&logo=github)](https:\u002F\u002Fgithub.com\u002FVoltAgent\u002Fawesome-agent-skills) · [![Project of the Day](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FAI%20Digital%20Crew-Project%20of%20the%20Day-orange?style=flat)](https:\u002F\u002Faidigitalcrew.com)\n\n- ⭐ **~10k GitHub stars** — top-trending AI agent repo\n- 🥇 **HuggingFace Daily Papers #1** — top of the day, paper [arXiv:2605.03042](https:\u002F\u002Fhuggingface.co\u002Fpapers\u002F2605.03042)\n- 🏆 **AI Digital Crew · Project of the Day** (2026.03.14)\n- 📰 **Featured on PaperWeekly** + **VoltAgent\u002Fawesome-agent-skills**\n- 🛠️ **74+ research skills** — full lifecycle from idea exploration → experiments → papers → rebuttals → talk slides\n- 🌐 **7+ platforms supported** — Claude Code · Codex CLI · Cursor · Trae · Antigravity · GitHub Copilot CLI · OpenClaw\n- 🔧 **ARIS-Code standalone CLI** — multi-provider runtime, no Claude Code dependency required\n\nCore methodology: **cross-model adversarial review** — executor and reviewer must come from different model families (Claude × GPT-5.5 xhigh × Gemini), so no LLM ever judges its own output. This protocol carries directly into interview cheat sheet generation: every formula, code block, and citation in every tutorial passes an independent audit (see each `.review.json` audit trail).\n\n👉 **ARIS main repo**: https:\u002F\u002Fgithub.com\u002Fwanshuiyin\u002FAuto-claude-code-research-in-sleep\n\n---\n\n## 📚 Tutorial Index\n\n> 🌐 **Bilingual editions**: every cheat sheet ships with both a Chinese (default) and an English HTML — filenames are `*_tutorial.html` (CN) and `*_tutorial_en.html` (EN). HTML columns below link to both.\n\n### 🧠 General \u002F Foundations\n\n| Topic | HTML 中文 | HTML EN | MD |\n|---|---|---|---|\n| **Attention Interview Cheat Sheet** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fattention_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fattention_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fattention_tutorial.md) |\n| **KL Divergence in RLHF (k1\u002Fk2\u002Fk3 · placement gradient bias)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fkl_divergence_rlhf_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fkl_divergence_rlhf_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fkl_divergence_rlhf_tutorial.md) |\n\n### 🎯 Post-Training & Reasoning\n\n| Topic | HTML 中文 | HTML EN | MD |\n|---|---|---|---|\n| **RLHF \u002F DPO \u002F GRPO \u002F PPO** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Frlhf_dpo_grpo_ppo_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Frlhf_dpo_grpo_ppo_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Frlhf_dpo_grpo_ppo_tutorial.md) |\n| **Reasoning Models (o1 \u002F R1 \u002F Test-Time Compute \u002F PRM)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Freasoning_models_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Freasoning_models_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Freasoning_models_tutorial.md) |\n| **LLM On-Policy Distillation (MiniLLM \u002F GKD \u002F Qwen3 \u002F Tinker)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fllm_opd_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fllm_opd_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fllm_opd_tutorial.md) |\n\n### 🏛️ LLM Architecture & Systems\n\n| Topic | HTML 中文 | HTML EN | MD |\n|---|---|---|---|\n| **MoE (DeepSeek-V3 \u002F Mixtral \u002F Llama 4)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fmoe_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fmoe_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fmoe_tutorial.md) |\n| **Long Context (RoPE \u002F YaRN \u002F NTK \u002F MLA \u002F StreamingLLM)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Flong_context_rope_yarn_mla_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Flong_context_rope_yarn_mla_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Flong_context_rope_yarn_mla_tutorial.md) |\n| **KV Cache + Speculative Decoding (Medusa \u002F EAGLE \u002F MLA)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fkv_cache_speculative_decoding_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fkv_cache_speculative_decoding_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fkv_cache_speculative_decoding_tutorial.md) |\n| **Quantization (GPTQ \u002F AWQ \u002F FP8 \u002F NVFP4 \u002F SmoothQuant)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fquantization_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fquantization_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fquantization_tutorial.md) |\n| **Distributed Training (DDP \u002F FSDP2 \u002F ZeRO \u002F TP \u002F PP \u002F EP \u002F SP)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fdistributed_training_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fdistributed_training_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fdistributed_training_tutorial.md) |\n\n### 🌊 Generative Models — Theory & Tokenizers\n\n| Topic | HTML 中文 | HTML EN | MD |\n|---|---|---|---|\n| **Flow Matching Quick Reference** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fflow_matching_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fflow_matching_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fflow_matching_tutorial.md) |\n| **Diffusion Foundations (DDPM \u002F Score \u002F DDIM \u002F EDM \u002F CFG)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fdiffusion_foundations_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fdiffusion_foundations_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fdiffusion_foundations_tutorial.md) |\n| **VAE \u002F VQ-VAE \u002F VQ-GAN \u002F FSQ** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fvae_vqvae_vqgan_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fvae_vqvae_vqgan_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fvae_vqvae_vqgan_tutorial.md) |\n\n### 🎨 Generation Systems — Image \u002F Video \u002F 3D \u002F Diffusion Post-Training\n\n| Topic | HTML 中文 | HTML EN | MD |\n|---|---|---|---|\n| **Image Gen Systems (LDM \u002F SD \u002F SDXL \u002F SD3 \u002F FLUX \u002F ControlNet)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fimage_generation_systems_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fimage_generation_systems_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fimage_generation_systems_tutorial.md) |\n| **Video Gen (Sora \u002F Hunyuan-Video \u002F Kling \u002F Wan \u002F Movie Gen)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fvideo_generation_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fvideo_generation_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fvideo_generation_tutorial.md) |\n| **3D Gen (NeRF \u002F Instant-NGP \u002F 3DGS \u002F SDS \u002F Trellis)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002F3d_generation_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002F3d_generation_tutorial_en.html) | [MD](docs\u002Ftutorials\u002F3d_generation_tutorial.md) |\n| **Diffusion Post-Training (DDPO \u002F DPOK \u002F DRaFT \u002F AlignProp \u002F Diffusion-DPO \u002F Flow-GRPO)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fdiffusion_post_training_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fdiffusion_post_training_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fdiffusion_post_training_tutorial.md) |\n| **Diffusion \u002F Flow Distillation (CM \u002F iCT \u002F sCM \u002F CTM \u002F LCM \u002F DMD\u002FDMD2 \u002F ADD\u002FLADD)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fdiffusion_distillation_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fdiffusion_distillation_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fdiffusion_distillation_tutorial.md) |\n\n### 👁️ Multimodal\n\n| Topic | HTML 中文 | HTML EN | MD |\n|---|---|---|---|\n| **VLM (CLIP \u002F LLaVA \u002F Qwen-VL \u002F DeepSeek-VL)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fvlm_multimodal_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fvlm_multimodal_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fvlm_multimodal_tutorial.md) |\n\n### 🤖 Agents\n\n| Topic | HTML 中文 | HTML EN | MD |\n|---|---|---|---|\n| **Agent Foundations (ReAct \u002F MCP \u002F A2A \u002F SWE-bench \u002F GAIA \u002F OSWorld)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fagent_foundations_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fagent_foundations_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fagent_foundations_tutorial.md) |\n| **Agentic RL (AgentTuning \u002F ToolRL \u002F RAGEN \u002F WebRL \u002F SWE-RL \u002F GRPO for tool use)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fagentic_rl_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fagentic_rl_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fagentic_rl_tutorial.md) |\n| **Multi-Agent & Long-Horizon (CAMEL \u002F AutoGen \u002F MetaGPT \u002F MoA \u002F Debate \u002F MemGPT \u002F LATS)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fmulti_agent_long_horizon_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fmulti_agent_long_horizon_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fmulti_agent_long_horizon_tutorial.md) |\n| **Self-Evolving Agents (Ctx2Skill \u002F Native Evolution \u002F A²RD \u002F Voyager \u002F Reflexion \u002F STaR)** | [📄 CN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fself_evolving_agents_tutorial.html) | [📄 EN](https:\u002F\u002Fwanshuiyin.github.io\u002FARIS-in-AI-Offer\u002Ftutorials\u002Fself_evolving_agents_tutorial_en.html) | [MD](docs\u002Ftutorials\u002Fself_evolving_agents_tutorial.md) |\n\n> 🎉 **23 tutorials live (bilingual)** (2026-05) — each ships with both Chinese and English HTML. Seven buckets: General · Post-Training · Architecture · Generative · Multimodal · Agents · Diffusion Post-Training. This round adds 4 new sheets: KL Divergence in RLHF, LLM On-Policy Distillation, Diffusion Post-Training, Diffusion Distillation. More (Flow-OPD \u002F Audio Gen \u002F further SOTA updates) coming — **PRs welcome** (see [CONTRIBUTING](CONTRIBUTING.md)).\n\n---\n\n## 🤖 How These Are Generated\n\nEvery tutorial uses ARIS's `\u002Finterview-cheatsheet` skill:\n\n1. **Plan** — 12-14 sections (TL;DR · Intuition · Formulas · Code · Variants · Complexity · 25 Q&A)\n2. **Draft** — 600-1000 lines of Chinese tutorial + runnable from-scratch PyTorch\n3. **Cross-model review** — fresh-thread codex GPT-5.5 xhigh audit on 10 properties (formula correctness · code runnability · citation accuracy · table-pipe escapes · callout style · personal-info leak · ...)\n4. **Fix loop** — trajectory-based; keep going if FAIL set is shrinking, stop if same issue recurs or ~6 rounds without convergence\n5. **`\u002Frender-html`** — single-file HTML render + 13-property render audit (information fidelity · TOC · math · code highlight · safety · privacy · ...)\n6. **`.review.json`** — full audit trail saved next to each tutorial\n\nCross-model adversarial review (executor ≠ reviewer family) is ARIS's core invariant: an LLM auditing its own output is no audit.\n\n---\n\n## 🚧 Coming features\n\n- 🌐 **ARIS-Homepage** *(TODO)* — auto-generate a personal portfolio \u002F academic homepage for 秋招 from your CV + project list + paper bibliography, powered by the same `\u002Frender-html` workflow that produces these cheat sheets. Single-file HTML, deploy-anywhere, MathJax + responsive layout, cross-model review on factual claims (paper venues, dates, model names) before publish. Designed for: PhD candidates putting together an academic homepage, intern candidates building a portfolio, anyone who wants to ship a polished personal site without fighting Hugo \u002F Jekyll \u002F Webflow. **Open an [issue](https:\u002F\u002Fgithub.com\u002Fwanshuiyin\u002FARIS-in-AI-Offer\u002Fissues\u002Fnew) if you'd like to shape the API.**\n\n---\n\n## 🤝 Contributing\n\nOne person can only cover so much. The hope is that many hands make this collection more complete.\n\nFull contribution guide: [**CONTRIBUTING.md**](CONTRIBUTING.md) ([English](CONTRIBUTING.md) · [中文](CONTRIBUTING_CN.md)) — covers ARIS workflow invocation, strict style guide (headings \u002F math \u002F tables \u002F callouts \u002F personal-info banlist), and PR checklist.\n\n**TL;DR**: use the [`\u002Finterview-cheatsheet`](skills\u002Finterview-cheatsheet\u002FSKILL.md) + [`\u002Frender-html`](skills\u002Frender-html\u002FSKILL.md) workflow to generate, then open a PR. Both skills enforce a cross-model codex GPT-5.5 xhigh review gate (math \u002F code \u002F citation \u002F render fidelity), so anything merged via PR has a baseline quality floor. Skill source and `tools\u002Frender_html.py` are bundled in this repo so you can fork & extend.\n\n**Honest disclaimer**: across the existing tutorials, the HTML structural foundations (math, code, tables, callouts, TOC, responsive layout) are solid. But the very latest frontier work in any given topic (e.g., methods released in late 2025, niche subfield updates) likely is not fully covered. If you spot something outdated or wrong, PRs and issues are equally welcome — let's keep this resource alive together.\n\n---\n\n## 💬 Community\n\n**Shared community with the main [ARIS repo](https:\u002F\u002Fgithub.com\u002Fwanshuiyin\u002FAuto-claude-code-research-in-sleep)** — the same WeChat group covers ARIS skill workflows + this tutorial collection. Join to discuss interview prep, request new cheat-sheet topics, or share corrections \u002F contributions:\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"assets\u002Fwechat_group.jpg\" alt=\"WeChat Group QR Code (shared with ARIS main repo)\" width=\"300\">\n\u003C\u002Fp>\n\n---\n\n## 📖 Citing ARIS\n\nIf this collection — or any cheat sheet here — helped you in your interview prep \u002F research \u002F paper, please consider citing the underlying ARIS methodology paper:\n\n```bibtex\n@article{yang2026aris,\n  title={ARIS: Autonomous Research via Adversarial Multi-Agent Collaboration},\n  author={Yang, Ruofeng and Li, Yongcan and Li, Shuai},\n  journal={arXiv preprint arXiv:2605.03042},\n  year={2026}\n}\n```\n\nEvery tutorial in this repo was generated end-to-end by the ARIS `\u002Finterview-cheatsheet` + `\u002Frender-html` workflow with cross-model adversarial review (Claude × GPT-5.5 xhigh × Gemini). The citation supports the methodology behind the workflow, not just this collection.\n\n---\n\n## License\n\n[MIT](LICENSE) — use, modify, share, fork freely. Hope this helps your job search. 💪\n","ARIS-in-AI-Offer 是一个为秋招准备的双语（中文+英文）机器学习面试速查手册，涵盖ML、LLM、多模态、扩散模型、代理和生成模型等内容。项目的核心功能是通过ARIS工作流自动生成结构化的面试指南，每份指南包括公式推导、从零开始的PyTorch代码实现以及25个高频面试问题（分为基础、进阶和顶级实验室三个层次）。这些内容以单文件HTML格式呈现，便于在手机、iPad或笔记本电脑上阅读。此项目适用于正在准备AI领域校园招聘的学生和技术人员，帮助他们更高效地复习关键知识点和面试技巧。","2026-06-11 03:55:54","CREATED_QUERY"]