Smooth upgrade from ranking to generative recommendation improves video engagement
LIGE-GR: A Smooth Leap from Ranking to Generative Recommendation in the LLM Era
Machine LearningInformation Retrieval
Summary
Recommender systems suggest videos or content you might like, usually by ranking items one by one. The authors propose a new method, LIGE-GR, that treats recommendation like generating a sequence, similar to how language models create sentences. This new approach works alongside existing systems without replacing them, making it easier to adopt. Tested on Instagram Reels and Facebook Video, it increased viewers’ time watching videos by a small but measurable amount using only a bit more computing power.
What this means in practice
- •For video platform engineers: Integrate a generative listwise recommendation approach to increase user engagement without overhauling existing ranking infrastructure.
- •For content recommendation system teams: Enhance mature recommendation stacks with listwise optimization techniques inspired by language model generation while maintaining compatibility.
Authors
Venkat Srinivas, Chenzhang He, Sam Woodmansee, Shawn Lian, Wenjie Hu, Renjie Jiang, Ziheng Huang, Xinyuan Zhang, Zhihao Zheng, Zhuoran Yu, Rui Li, Lei Yuan, Ziwei Li, Jimmy Jia, Mert Terzihan, Ekrem Kocaguneli, Yiming Liao, Zhichen Zhao, Yue Yin, Yue Weng, Wanlin Ma, Xufeng Cai, Weimiao Wu, Yezhou Huang, Du Zhang, Yukun Ding, Aaron Johnston, Yueming Wang, Zhaojie Gong, Yuting Zhang, Serena Li, Adithya Ganesh, Boying Liu, Haichuan Yang, Xialu Li, Matt Ma, Qunshu Zhang, John Joshua Miller, Praveen Rathinavelu, Cheng Huang, Aadhar Sachdeva, Josh Karns, Andres Aaron Gutierrez, Neil Agarwal, Gustas Pladis, Vladimir Batygin, Gopal Ray, Aditya Priyadarshi, Shantanu Patil, Zhe Wang, Penny Pan, Yiping Han, Arun Singh, Guangdeng Liao, Bi Xue, Xinyao Hu, Yang Song, Yisong Song, Meihong Wang, Haotian Wu, Deepak Agarwal, Ji Liu
Abstract
The remarkable success of large language models (LLMs) has provided important inspiration for the next generation of recommender systems. Structurally, recommendation and language generation share a similarity: both aim to produce an ordered sequence that optimizes the user's experience. However, how to precisely absorb the essence of the LLM paradigm into mature industrial recommender systems remains an open problem. There are two challenges. First, it is unclear how to incorporate sequence-level generation and optimization from the LLM paradigm into recommendation. Second, real-world recommender systems are mature systems that have been iteratively customized for years around specific products, business constraints, serving infrastructure, and organizational ownership. Replacing such systems wholesale is often technically risky and organizationally disruptive. In this paper, we propose LIGE-GR, a listwise generation and evaluation recommendation framework that upgrades from a traditional ranking system based on itemwise recommendation toward a generative recommendation paradigm. Instead of rebuilding the entire recommendation stack from scratch, LIGE-GR generalizes the existing pointwise recommendation system into a listwise generation system. This allows mature recommender systems to benefit from listwise optimization while preserving compatibility with existing models, value functions, and serving infrastructure. We validate LIGE-GR in short-video recommendation on Instagram Reels and Facebook Video. On these recommendation surfaces, LIGE-GR improves time spent by 1.14 percent on Instagram Reels and 0.72 percent on Facebook Video, while requiring only modest additional inference resources.