11 papers
Lightweight Ranking Heads: Accelerating Multi-Task Experimentation in Production Recommender Systems
Sanjay Surendranath Girija, Aniruddh Nath, Li Wei +6
Modern production-scale recommender systems rely on complex, multi-task ranking models. Introducing new prediction tasks into these massive systems often causes bottlenecks - it ri…
Code-to-Harness: Distilling Black-Box Optimizers from Self-Play
Yi Wu, Zheng Ren, Zhiyu Hu +8
Can an agent learn a numerical search strategy through executable practice and then transfer that strategy as text? We study low-budget black-box optimization, where unaided langua…
Token Factory: Efficiently Integrating Diverse Signals into Large Recommendation Models
Xilun Chen, Shao-Chuan Wang, Baykal Cakici +6
Large Recommendation Models (LRMs) have demonstrated promising capabilities in industry-scale recommendation tasks. However, holistically integrating traditional signals into these…
TokenMinds: Pretrained User Tokens and Embeddings for User Understanding in Large Recommender Systems
Qingyun Liu, Bo Yan, Yang Liu +16
User modeling in industrial recommender systems typically produces dense embeddings, which suffer from representational constraints due to fixed-dimensional vectors. As an emerging…
ORBIT: Preserving Foundational Language Capabilities in GenRetrieval via Origin-Regulated Merging
Neha Verma, Nikhil Mehta, Shao-Chuan Wang +7
Despite the rapid advancements in large language model (LLM) development, fine-tuning them for specific tasks often results in the catastrophic forgetting of their general, languag…
Learning to Alleviate Familiarity Bias in Video Recommendation
Zheng Ren, Yi Wu, Jianan Lu +4
Modern video recommendation systems aim to optimize user engagement and platform objectives, yet often face structural exposure imbalances caused by behavioral biases. In this work…
Vectorizing the Trie: Efficient Constrained Decoding for LLM-based Generative Retrieval on Accelerators
Zhengyang Su, Isay Katsman, Yueqi Wang +10
Generative retrieval has emerged as a powerful paradigm for LLM-based recommendation. However, industrial recommender systems often benefit from restricting the output space to a c…
Self-Evolving Recommendation System: End-To-End Autonomous Model Optimization With LLM Agents
Haochen Wang, Yi Wu, Daryl Chang +2
Optimizing large-scale machine learning systems, such as recommendation models for global video platforms, requires navigating a massive hyperparameter search space and, more criti…
Selecting User Histories to Generate LLM Users for Cold-Start Item Recommendation
Nachiket Subbaraman, Jaskinder Sarai, Aniruddh Nath +4
Large Language Models (LLMs) have demonstrated remarkable capabilities in reasoning, generalization, and simulating human-like behavior across a wide range of tasks. These strength…
ACT: Automated Constraint Targeting for Multi-Objective Recommender Systems
Daryl Chang, Yi Wu, Jennifer She +2
Recommender systems often must maximize a primary objective while ensuring secondary ones satisfy minimum thresholds, or "guardrails." This is critical for maintaining a consistent…
Learned Ranking Function: From Short-term Behavior Predictions to Long-term User Satisfaction
Yi Wu, Daryl Chang, Jennifer She +3
We present the Learned Ranking Function (LRF), a system that takes short-term user-item behavior predictions as input and outputs a slate of recommendations that directly optimizes…