Contemporary recommendation systems are designed to meet users' needs by delivering tailored lists of items that align with their specific demands or interests. In a multi-stage recommendation system, reranking plays a crucial role by modeling the intra-list correlations among items. The key challenge of reranking lies in the exploration of optimal sequences within the combinatorial space of permutations. Recent research proposes a generator-evaluator learning paradigm, where the generator generates multiple feasible sequences and the evaluator picks out the best sequence based on the estimated listwise score. The generator is of vital importance, and generative models are well-suited for the generator function. Current generative models employ an autoregressive strategy for sequence generation. However, deploying autoregressive models in real-time industrial systems is challenging. To address these issues, we propose a Non-AutoRegressive generative model for reranking Recommendation (NAR4Rec) designed to enhance efficiency and effectiveness. To tackle challenges such as sparse training samples and dynamic candidates, we introduce a matching model. Considering the diverse nature of user feedback, we employ a sequence-level unlikelihood training objective to differentiate feasible sequences from unfeasible ones. Additionally, to overcome the lack of dependency modeling in non-autoregressive models regarding target items, we introduce contrastive decoding to capture correlations among these items. Extensive offline experiments validate the superior performance of NAR4Rec over state-of-the-art reranking methods. Online A/B tests reveal that NAR4Rec significantly enhances the user experience. Furthermore, NAR4Rec has been fully deployed in a popular video app Kuaishou with over 300 million daily active users.
Efficient Personalized Reranking with Semi-Autoregressive Generation and Online Knowledge Distillation
University of Science and Technology of China（中国科学技术大学）；Huawei（华为）；Shanghai Jiao Tong University（上海交通大学）；Beijing University of Posts and Telecommunications（北京邮电大学）
DeGRe: Dense-supervised Generative Reranking for Recommendation
College of Software, Zhejiang University Hangzhou China；Rajax Network Technology, Taobao Shangou of Alibaba Hangzhou China；Rajax Network Technology, Taobao Shangou of Alibaba Beijing China；State Key Lab of CAD\&CG, Zhejiang University Hangzhou China；Rajax Network Technology, Taobao Shangou of Alibaba Shanghai China；College of Software, Zhejiang University；Rajax Network Technology, Taobao Shangou of Alibaba；State Key Lab of CAD\&CG, Zhejiang University
LLM-Assisted Reranking to Operationalize Nuanced Objectives in Recommender Systems
Department of Communication, University of California, Los Angeles, CA 90095（通信系，加州大学洛杉矶分校，CA 90095）；Department of Computer and Information Science, University of Pennsylvania, Philadelphia, PA 19104（计算机与信息科学系，宾夕法尼亚大学，Philadelphia, PA 19104）；Amenenberg School of Communication, University of Pennsylvania, Philadelphia, PA 19104（安纳伯格通信学院，宾夕法尼亚大学，Philadelphia, PA 19104）；Operations, Information, and Decisions Department, University of Pennsylvania, Philadelphia, PA 19104（运营、信息与决策系，宾夕法尼亚大学，Philadelphia, PA 19104）
Revisiting Reciprocal Recommender Systems: Metrics, Formulation, and Method
Nanbeige Lab（南戈实验室）；BOSS Zhipin（BOSS直聘）；Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）；University of California, San Diego（加利福尼亚大学圣迭戈分校）；Career Science Lab, BOSS Zhipin（BOSS直聘职业科学实验室）
Fairness for niche users and providers: algorithmic choice and profile portability
Department of Information Science, University of Colorado, Boulder（信息科学系，科罗拉多大学，伯尔德）；Department of Computer Science, University of Colorado, Boulder（计算机科学系，科罗拉多大学，伯尔德）
Give Users the Wheel: Towards Promptable Recommendation Paradigm
McGill \& Mila - Quebec AI Institute Montreal Canada；Shenzhen Technological University Shenzhen China；University of California San Diego San Diego US；McGill University Montreal Canada；McGill \& Mila - Quebec AI Institute；Shenzhen Technological University；University of California San Diego；McGill University
