works on

From the 3 of 7 linked papers with an AI index.

collaborators

7 papers

cs.IR2026

Cascading Relevance-driven Recommendation Network for CTR Prediction in Trigger-Introduced Recommendation

Kaixuan Chen, Wenwen Wang, Xing Fang +2

E-commerce has emerged as crucial platforms for people's daily consumption and shopping interests. There is a new recommendation scenario, Trigger-Introduced Recommendation (TIR),…

cs.IR2026

TMallGS: Scaling Unified Feature and Sequence Modeling for Generative E-commerce Search

Zhentao Song, Yufeng Gao, Xing Fang +5

The paper introduces TMallGS, a transformer-based ranking architecture for e‑commerce search that combines specialized tokenization, field‑adaptive transformers, and bias‑aware tra…

cs.IR2026

Learning to Forget: Satiation-Aware Long-Sequence Transducers for Mitigating Post-Purchase Redundancy

Yipin Dai, Ruocong Tang, Xing Fang +4

The paper introduces a satiation-aware framework for sequential recommendation that detects when a purchase satisfies a user’s intent and temporarily suppresses related items, then…

cs.IR2026

Cheaper is Better: A Discount-Aware Network for Conversion Rate Prediction in E-commerce Recommendation System

Ruocong Tang, Yang Huang, Xing Fang +3

The paper introduces a Discount-Aware Network (DANet) that incorporates item discount information via Fourier-based time‑frequency analysis and bias‑mitigation modules to improve p…

cs.IR2026

DSIRM: Learning Query-Bridged Discrete Semantic Identifiers for E-commerce Relevance Modeling

Bokang Wang, Xing Fang, Mingmin Jin +4

Despite rapid progress of continuous embeddings for e-commerce search relevance, a long-standing open problem is the difficulty in capturing fine-grained attribute distinctions. Wh…

cs.IR2026

From Head to Tail: Asymmetric Knowledge Transfer in Long-tail Recommendation with Generative Semantic IDs

Chenyi Yan, Ruocong Tang, Xing Fang +3

Long-tail recommendation in real-world e-commerce platforms remains challenging due to severe data imbalance. Existing methods often struggle to combine content-based multimodal fe…