arXiv AI By Tianqi Liu, Kaipu Sun, Kairui Fu, Wenkai Wang, Sihao Liu, Hui Zang, Ran Chen, Tao Wang, Shengyu Zhang

RACE: Ranking-Uncertainty Aware Capacity-Adaptive Customization for Efficient Sequential Recommendation

Read the original on arXiv AI →

RACE is a framework for efficient sequential recommendation that adapts model structure and input sequences to each instance’s complexity. It uses a ranking‑uncertainty metric derived from the stability of top‑K predictions to estimate instance difficulty, then predicts this uncertainty from early‑layer features. Guided by the prediction, RACE performs instance‑specific slimming—channel sensitivity extraction to remove redundant parameters and progressive sequence filtering to reduce data redundancy—achieving better accuracy, efficiency, and adaptability than existing baselines on four datasets.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI
Sep 10

SequenceO1: End-to-End Ultra-Long (100K) Sequence Modeling in Recommendation with Low-Rank Caching

SequenceO1 is an end‑to‑end framework that enables ultra‑long (up to 100K interactions) sequence modeling for recommendation systems. It compresses raw user histories into a fixed‑size sketch using Sketch Attention and then models short‑term and long‑term interests with Target‑to‑History Cross Attention. The system incorporates low‑rank caching, batching, pipeline lift, and a FlashSA kernel to keep training and inference efficient, achieving consistent offline and online performance gains when deployed at full traffic on Douyin.

By Lin Guan, Jia-Qi Yang, Zhishan Zhao, Jiaqi Huang, Hangyu Wang, Longbin Li, Beichuan Zhang, Haonan Jiang, Jinan Ni, Xiangyu Fan, Xiaowen Li, Ziyao Ren, Yuhang Qi, Xiaolong Zhu, Xuanyuan Luo, Qiwei Chen, Yi Cheng, Lele Yu
arXiv Machine Learning
Jul 31

ROCS: Request-Oriented Compute Sharing for Efficient Large-Scale Recommendation

arXiv:2607. 27744v1 Announce Type: new Abstract: Modern recommendation models gain prediction quality by scaling feature-interaction and sequence modules, but production cost constraints cap how far systems can scale.

By Yuxin Chen, Liang Luo, Buyun Zhang, Jian Jiao, Boda Li, Haoyu Wang, Tongyi Tang, Ao Cai, Zijian Shen, Zhengkai Zhang, Wenyi Xie, Ryan Dick, Han Liu, Neng Shi, Bin Yu, Jianbo Xiao, Shuyao Bi, Hongtao Yu, Yuanwei Fang, Zhuoran Zhao, Sijia Chen, Yang Chen, Shuqi Yang, Qianru Li, Zikun Liu, Wei Ling, Sihan Zeng, Longhao Jin, Jiaxin Lu, Yinbin Ma, Jiawei Li, Yichen Ruan, Yong Ler Lee, Birmingham Guan, Zijian Li, Jianbo Sun, Zhengyu Zhang, Zeliang Chen, Xiaohan Wei, Yuchen Hao, GP Musumeci, Venkatesh Ranganathan, Yantao Yao, Chunqiang Tang, Wenlin Chen, Santanu Kolay, Ellie Dingqiao Wen
arXiv AI
Jun 12

Versioned Late Materialization for Ultra-Long Sequence Training in Recommendation Systems at Scale

arXiv:2604. 24806v2 Announce Type: replace-cross Abstract: Modern Deep Learning Recommendation Models (DLRMs) follow scaling laws with sequence length, driving the frontier toward ultra-long User Interaction History (UIH).

By Liang Guo, Ge Song, Litao Deng, Jianhui Sun, Chufeng Hu, Lu Zhang, Zhen Ma, Shouwei Chen, Weiran Liu, Sarang Masti Sreeshylan, Xiaoxuan Meng, Yanzun Huang
arXiv AI
Nov 2

LEAP: Efficient LLM Inference Scheduling with Generalizable Length Prediction

LEAP is a length prediction system for large language model inference that uses the serving model’s prefill hidden states and a linear readout to estimate output length. It adapts online via recursive least squares, updating the readout with completed requests without costly retraining or gradient computation. Experiments show LEAP maintains strong ranking performance across domain shifts, improves Spearman correlation by 64.7% under large shifts, and reduces per-token latency by 15.9–40.8% while increasing throughput and drastically cutting predictor parameters.

By Kwan Yan, Jialun Zhong, Yanzeng Li
arXiv AI
Sep 25

DeGRe: Dense-supervised Generative Reranking for Recommendation

DeGRe is a dense‑supervised generative reranking framework designed to improve multi‑stage recommender systems by addressing label bias and credit assignment issues. It uses an offline Lookahead Evaluator with beam search to generate dense supervision signals, which are distilled into a lightweight Online Generator that can perform efficient greedy decoding at inference time. Experiments show that DeGRe outperforms baselines on public benchmarks and industrial datasets, and it has been successfully deployed on Taobao Flash Shopping to enhance online recommendations.

By Chaotian Song, Jingyao Zhang, Chenghao Chen, Zisen Sang, Dehai Zhao, Guodong Cao, Boxi Wu, Deng Cai, Jia Jia