5 papers
Generative Recommendation for Large-Scale Advertising
Ben Xue, Dan Liu, Lixiang Wang +27
Generative recommendation has recently attracted widespread attention in industry due to its potential for scaling and stronger model capacity. However, deploying real-time generat…
KAT-Coder-V2 Technical Report
Fengxiang Li, Han Zhang, Haoyang Huang +43
We present KAT-Coder-V2, an agentic coding model developed by the KwaiKAT team at Kuaishou. KAT-Coder-V2 adopts a "Specialize-then-Unify" paradigm that decomposes agentic coding in…
AlignGR: Unified Multi-Level Alignment for LLM-based Generative Recommendation
Wencai Ye, Mingjie Sun, Shuhang Chen +2
Large Language Models (LLMs) demonstrate significant advantages in leveraging structured world knowledge and multi-step reasoning capabilities. However, fundamental challenges aris…
DAS: Dual-Aligned Semantic IDs Empowered Industrial Recommender System
Wencai Ye, Mingjie Sun, Shaoyun Shi +3
Semantic IDs are discrete identifiers generated by quantizing the Multi-modal Large Language Models (MLLMs) embeddings, enabling efficient multi-modal content integration in recomm…
SweetTok: Semantic-Aware Spatial-Temporal Tokenizer for Compact Video Discretization
Zhentao Tan, Ben Xue, Jian Jia +7
This paper presents the \textbf{S}emantic-a\textbf{W}ar\textbf{E} spatial-t\textbf{E}mporal \textbf{T}okenizer (SweetTok), a novel video tokenizer to overcome the limitations in cu…