A method to fix GPT-3 after deployment with user feedback, without re-training.

Code for the paper "Aligning LLM Agents by Learning Latent Preference from User Edits".

Code for RecSys'19 paper: Leveraging Post-click Feedback for Content Recommendations

This web app recommends trending movies and songs by analyzing real-time tweets using sentiment analysis. It fetches data from Spotify and TMDB, and sends personalized email updates to users.

This is the repository for the paper "Learning from Implicit User Feedback, Emotions and Demographic Information in Task-Oriented and Document-Grounded Dialogues", which appears in EMNLP 2024 Findings.

GEO Knowledge Hub comments module