Llm

This post explores how DeepSeek R1 improves upon DeepSeek R1 Zero by addressing key challenges through strategic fine-tuning and reinforcement learning.

A deep dive into DeepSeek R1 Zero - exploring its architecture, training methodology, and breakthrough achievements in reasoning capabilities

summarization of the prompt engineering paper - Faithful Chain of Thought Reasoning by Lyu et al.

This post outlines key prompt engineering techniques to optimize LLM performance and improve model accuracy.