A framework for real-time LLM-based user interest personas in large-scale video recommendations, using distillation, async inference, and video clustering to balance interests with novel topics and improve viewer value via A/B tests.
Metadata generation and evaluation using llms - case study on canonical titles,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
An ablation study on 252 datasets finds that adding table schemas to LLM prompts consistently degrades narrative quality of generated descriptions compared to titles alone.
citing papers explorer
-
LLM-Based User Personas for Recommendations at Scale
A framework for real-time LLM-based user interest personas in large-scale video recommendations, using distillation, async inference, and video clustering to balance interests with novel topics and improve viewer value via A/B tests.
-
Less Is More? When Dataset Context Hurts LLM-Generated Dataset Descriptions
An ablation study on 252 datasets finds that adding table schemas to LLM prompts consistently degrades narrative quality of generated descriptions compared to titles alone.