[TOC]

  1. Title: Code Llama Open Foundation Model for Code
  2. Author: Baptiste Roziere et. al. META AI
  3. Publish Year: 2023
  4. Review Date: Mon, Oct 16, 2023
  5. url: https://scontent.fmel13-1.fna.fbcdn.net/v/t39.2365-6/369856151_1754812304950972_1159666448927483931_n.pdf?_nc_cat=107&ccb=1-7&_nc_sid=3c67a6&_nc_ohc=Hcg6QsYJx1wAX_okEZO&_nc_ht=scontent.fmel13-1.fna&oh=00_AfAYtfHJfYeomAQWiMUTRo96iP8d4sZrlIfD_KAeYlYaDQ&oe=6531E8CF

Summary of paper

image-20231016025929381

Motivation

Contribution

image-20231016032328670

Some key terms

code-training from foundation models

infilling

image-20231016032428930

instruction fine-tuning

image-20231016032716835

Dataset

prefix-suffix-middle (PSM)

image-20231016034857286

image-20231016034950496

self-instruct

Major comments

NO information about back-translation objective in this paper.