AI, But Simple

AI, But Simple

AI Research, Simplified. Weekly visual explanations to understand the latest research in minutes.

Connect

Featured Posts

The Dragon Hatchling (BDH), Simply Explained

The Dragon Hatchling (BDH), Simply Explained

AI, But Simple Issue #105

Edwin Dong
Edwin Dong
A Deep Dive into Modern Transformers: RoPE, Activation Functions, Quantization

A Deep Dive into Modern Transformers: RoPE, Activation Functions, Quantization

AI, But Simple Issue #86

Edwin Dong
Edwin Dong
A Deep Dive into Modern Transformers: Optimizing Attention

A Deep Dive into Modern Transformers: Optimizing Attention

AI, But Simple Issue #85

Edwin Dong
Lalit Julapalli
Edwin Dong, +1
Small Language Models (The Future?)

Small Language Models (The Future?)

AI, But Simple Issue #68

Edwin Dong
Edwin Dong
Quantization for LLMs, Simply Explained

Quantization for LLMs, Simply Explained

AI, But Simple Issue #61

Edwin Dong
Edwin Dong
Mamba and State Space Models, Simply Explained

Mamba and State Space Models, Simply Explained

AI, But Simple Issue #58

Edwin Dong
Edwin Dong

Archive

Context Self-Management, Generation for Understanding, Disaggregated Quantization

Context Self-Management, Generation for Understanding, Disaggregated Quantization

AI, But Simple Issue #122

Edwin Dong
Edwin Dong
Claude Opus 5.5, Agent Swarms Writing Proofs, Cheap Sparse Models

Claude Opus 5.5, Agent Swarms Writing Proofs, Cheap Sparse Models

AI, But Simple Issue #121

Edwin Dong
Edwin Dong
Recursive Self-Improvement, Ternary 27B Model for Laptops, Jev Architecture

Recursive Self-Improvement, Ternary 27B Model for Laptops, Jev Architecture

AI, But Simple Issue #120

Edwin Dong
Edwin Dong
LLMsLLMs
+1+1
Random Attention, Flow Matching for VLAs, Tencent's Speech Foundation Model

Random Attention, Flow Matching for VLAs, Tencent's Speech Foundation Model

AI, But Simple Issue #119

Edwin Dong
Edwin Dong
RL Training Inside Harnesses, Stolen Reasoning Traces, Judging Agents

RL Training Inside Harnesses, Stolen Reasoning Traces, Judging Agents

AI, But Simple Issue #118

Edwin Dong
Edwin Dong
How DeepSeek’s Architecture Actually Works

How DeepSeek’s Architecture Actually Works

AI, But Simple Issue #117

Edwin Dong
Lalit Julapalli
Edwin Dong, +1
Persistent AI memory, AI watermarking, on-policy distillation

Persistent AI memory, AI watermarking, on-policy distillation

AI, But Simple Issue #116

Edwin Dong
Edwin Dong
ArchitectureArchitecture
BDH-CQ: A 150M model just broke the ARC-AGI cost curve

BDH-CQ: A 150M model just broke the ARC-AGI cost curve

AI, But Simple Issue #115

Edwin Dong
Edwin Dong
LLMsLLMs
A Simple Guide to LLM Activation Steering With Control Systems

A Simple Guide to LLM Activation Steering With Control Systems

AI, But Simple Issue #114

Anurag Shinde
Edwin Dong
Anurag Shinde, +1
PapersPapers
The Best ICML Papers of 2026, Simply Explained

The Best ICML Papers of 2026, Simply Explained

AI, But Simple Issue #113

Edwin Dong
Edwin Dong
(PRO) Sparse Autoencoders (SAEs), Simply Explained
Premium

(PRO) Sparse Autoencoders (SAEs), Simply Explained

AI, But Simple Issue #112

Edwin Dong
Lalit Julapalli
Edwin Dong, +1
Sparse Autoencoders (SAEs), Simply Explained

Sparse Autoencoders (SAEs), Simply Explained

AI, But Simple Issue #112

Edwin Dong
Lalit Julapalli
Edwin Dong, +1
FirstBack
12345678
Next Last