ball-blog
  • Welcome, I'm ball
  • Machine Learning - Basic
    • Entropy
    • Cross Entropy
    • KL-Divergence
    • Monte-Carlo Method
    • Variational Auto Encoder
    • SVM
    • Adam Optimizer
    • Batch Normalization
    • Tokenizer
    • Rotary Positional Encoding
    • Vector Quantized VAE(VQ-VAE)
    • DALL-E
    • Diffusion Model
    • Memory Layers at Scale
    • Chain-of-Thought
  • Einsum
  • Linear Algebra
    • Linear Transformation
    • Determinant
    • Eigen-Value Decomposition(EVD)
    • Singular-Value Decomposition(SVD)
  • AI Accelerator
    • CachedAttention
    • SGLang
    • CacheBlend
  • Reinforcement Learning
    • Markov
  • Policy-Improvement Algorithm
  • Machine Learning - Transformer
    • Attention is All you need
    • Why do we need a mask in Transformer
    • Linear Transformer
    • kr2en Translator using Tranformer
    • Segment Anything
    • MNIST, CIFAR10 Classifier using ViT
    • Finetuning PaliGemma using LoRA
    • LoRA: Low-Rank Adaptation
  • EGTR: Extracting Graph from Transformer for SGG
  • Machine Learning - Mamba
    • Function Space(Hilbert Space)
    • HIPPO Framework
    • Linear State Space Layer
    • S4(Structures Space for Sequence Model)
    • Parallel Scan Algorithm
    • Mamba Model
  • Computer System
    • Memory Ordering: Release/Acquire 1
    • Memory Ordering: Release/Acquire 2
    • BUDAlloc
    • Lock-free Hash Table
    • Address Sanitizer
  • App development
    • Bluetooth connection in linux
    • I use Bun!
    • Using Tanstack-query in Frontend
    • Deploying AI Service using EC2
  • Problem Solving
    • Bipartite Graph
    • Shortest Path Problem in Graph
    • Diameter of a Tree
  • Scribbles
Powered by GitBook
On this page
  • Summary
  • References

Was this helpful?

Edit on GitHub
  1. Machine Learning - Basic

Chain-of-Thought

Last updated 3 months ago

Was this helpful?

Summary

Due to scaling law of AI Model, LLMs are getting bigger and bigger. However, for reasoning tasks, performance improvement was poor.

Researchers came up with an idea of instead of changing the model checkpoints, they tried to make the model mimic human's thinking process using prompt engineering.

Chain-of-Thought technique is making a model use a reasoning-chain like human.

Instead of making model just output the answer, we can make the model output all the reasoning path approaching to the answer. Since LLMs(Transformer) are auto-regressive models, it reference the previous reasoning path to approach to the answer.

References

[1]

[2]

https://arxiv.org/abs/2201.11903
https://arxiv.org/abs/2205.11916
LogoLarge Language Models are Zero-Shot ReasonersarXiv.org
Comparing Standard and CoT. From
https://arxiv.org/abs/2205.11916
LogoChain-of-Thought Prompting Elicits Reasoning in Large Language ModelsarXiv.org