TamSiuhin/PerRecBench
Official Implementation of "Can Large Language Models Understand Preferences in Personalized Recommendation?"
This project helps evaluate how well large language models (LLMs) truly understand individual user preferences for recommendations. It takes in user interaction data (like past ratings or interactions) and assesses the LLM's ability to recommend items based purely on preference, minimizing the influence of general item popularity or typical user rating habits. This is useful for product managers or researchers building recommendation systems who want to understand the true personalization capabilities of LLMs.
Use this if you are building or evaluating a personalized recommendation system using large language models and want to accurately measure how well it captures individual user preferences, beyond just predicting high ratings.
Not ideal if you are looking for a complete, production-ready recommendation system, as this project focuses specifically on evaluating preference understanding rather than providing an end-to-end solution.
Stars
11
Forks
1
Language
Python
License
—
Category
Last pushed
Feb 06, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/TamSiuhin/PerRecBench"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
AkaliKong/MiniOneRec
Minimal reproduction of OneRec
microsoft/RecAI
Bridging LLM and Recommender System.
dokar3/upnext-gpt
GPT powered playlist App for Android. Supports Apple Music, Spotify, and Youtube Music.
YuanchenBei/Awesome-Cold-Start-Recommendation
[Up-to-date] A curated list of resources on cold-start recommendations.
giuseppe99barchetta/SuggestArr
Effortlessly request recommended movies, TV shows and anime to Jellyseer/Overseer based on your...