SOURCES
Teaching Gemini to Speak YouTube: Adapting LLMs for Video Recommendations
Deep Neural Networks for YouTube Recommendations
GIST: Greedy Independent Set Thresholding for Max-Min Diversification with Submodular Utility
PLUM: Adapting Pre-trained Language Models for Industrial-scale Generative Recommendations
PATENT APPLICATION: SEARCH WITH GENERATIVE ARTIFICIAL INTELLIGENCE -Ā US 2024/0256582 A1
PATENT: CONTEXTUAL ESTIMATION OF LINK INFORMATION GAIN -Ā Patent No: US 11,354,342 B2Ā
Maximum Diversity Problem. A Multi-Objective Approach
Introducing GIST: The next stage in smart sampling
Recommender Systems with Generative Retrieval
Enhancing Subsequent Video Retrieval via Vision-Language Models (VLMs)
Training an LLM-RecSys Hybrid for Steerable Recs with Semantic IDs
Better Generalization with Semantic IDs: A Case Study in Ranking for Recommendations
YouTube gets ~5% CTR lift on Shorts by replacing embedding tables with Semantic IDs
John ScottĀ
YouTube Expert / Attention HackerĀ
