---
language: 
  - en

tags:
- sequential-recommendation
- direct-recommendation
- explanation-generation
- text2text-generation
  
license: mit

datasets:
- amazon_us_reviews

metrics:
- NDCG
- HR
- MAE
- BLUE
- ROUGE

widget:
- text: "According to the purchase history of JSB727 : \n 2440 -> 1212 -> 4234 -> 4309 \n Can you recommend the next possible item to the user ?"
  example_title: "Sequential Recommendation"
- text: "Pick the most suitable item from the following list and recommend to user_250 : \n 4915 , 1823 , 3112 , 3821 , 3773 , 520 , 7384 , 7469 , 9318 , 3876 , 1143 , 789 , 595 , 3824 , 3587"
  example_title: "Direct Recommendation"
- text: "Based on the feature word kids , generate an explanation for user_12107 about this product : Tumblin' Monkeys Game"
  example_title: "Explanation Generation"
---

# P5 (Toys Small)
Recommendation as Language Processing: A Unified Pretrain, Personalized Prompt & Predict Paradigm 

![model image](https://raw.githubusercontent.com/jeykigung/P5/main/pic/teaser.png)