The big picture

Think of building a modern language model as a four-step journey. Each paper in this folder improves one step.

flowchart TD
    A[Step 1: Architecture<br/>What shape is the model?] --> B[Step 2: Scale<br/>How big, how much data?]
    B --> C[Step 3: Alignment<br/>How do we make it helpful?]
    C --> D[Step 4: Efficiency<br/>How do we train and run it cheaply?]
    D --> E[Step 5: Evaluation<br/>How do we know it is good?]

Here is how the eight papers map onto that journey.

Step	Question it answers	Papers
Architecture	What internal design lets a model understand language?	Attention Is All You Need
Scale	How big should the model and dataset be?	Scaling Laws, Chinchilla
Alignment	How do we turn a raw text predictor into a helpful assistant?	InstructGPT (RLHF), DPO
Efficiency	How do we customize and serve models without huge cost?	LoRA, Mixtral (Mixture of Experts)
Evaluation	How do we measure whether a model is actually good?	LLM-as-a-Judge / Chatbot Arena