DuoLM: Fast LLM-Based Personalized Recommendation via Compact Encoder-Ranker
GoogleDeepMind · hf · 2026-10-26
Researchers from Google DeepMind, UC Irvine and Oracle propose DuoLM, a framework for fast and effective personalized recommendation with language models.
- Problem: scoring many candidates with full-text LLM inputs incurs heavy online computation and limits serving throughput
- Approach: an encoder LLM learns compact item representations offline for reuse at runtime, paired with a compact ranker and optimized inference
- Achieves high throughput with minimal quality loss
- The paper also discusses why traditional ID-based two-tower recommenders trained only on interactions fall short on content understanding
More from Research
- Judea Pearl boosts Oregon State PhD opening at causal inference × AI intersection — yudapearl · 2026-10-26
- HiP3R Stitches Long Video Into One Coherent 3D Scene Across Thousands of Frames — ccloy · 2026-10-26
- davidad digs into 'dynamical-systems attractor-structure functionalism' via Conley index theory — davidad · 2026-10-26
- Sakana AI hosts Tokyo RSI symposium: AI Scientist in Nature, Schmidhuber keynotes — RobertTLange · 2026-10-26
- HuggingFace's Carbon-X Spots Pathogenic Splice Variant Where Gemini 3.8 Flash Misreads the Gene — LoubnaBenAllal1 · 2026-10-26
- ETH Zurich paper: dynamic evaluation allocation matches 10x-budget model rankings — zouharvi · 2026-10-26