4 papers
MemRerank: Preference Memory for Personalized Product Reranking
Zhiyuan Peng, Xuyang Wu, Huaixiao Tou +2
LLM-based shopping agents increasingly rely on long purchase histories and multi-turn interactions for personalization, yet naively appending raw history to prompts is often ineffe…
An Accelerated Proximal Bundle Method with Momentum
Zhuoqing Zheng, Junshan Yin, Shaofu Yang +1
Proximal bundle methods (PBM) are a powerful class of algorithms for convex optimization. Compared to gradient descent, PBM constructs more accurate surrogate models that incorpora…
Bundle EXTRA for Decentralized Optimization
Haijuan Liu, Zhuoqing Zheng, Cong Li +2
Decentralized primal-dual methods are widely used for solving decentralized optimization problems, but their updates often rely on the potentially crude first-order Taylor approxim…
From promise to practice: realizing high-performance decentralized training
Zesen Wang, Jiaojiao Zhang, Xuyang Wu +1
Decentralized training of deep neural networks has attracted significant attention for its theoretically superior scalability over synchronous data-parallel methods like All-Reduce…