3 papers
cs.CV2025
When Deepfake Detection Meets Graph Neural Network:a Unified and Lightweight Learning Framework
Haoyu Liu, Chaoyu Gong, Mengke He +3
The proliferation of generative video models has made detecting AI-generated and manipulated videos an urgent challenge. Existing detection approaches often fail to generalize acro…
cs.DB2025
On 10x Better Scalability: KV Stores Scale Up KV Cache
Weiping Yu, Ye Jiarui, He Mengke +2
Large language models (LLMs) rely on Key-Value (KV) cache to reduce time-to-first-token (TTFT) latency, but existing disk-based KV cache systems using file-per-object layouts suffe…
cs.IR2025
Right Answer at the Right Time - Temporal Retrieval-Augmented Generation via Graph Summarization
Zulun Zhu, Haoyu Liu, Mengke He +1
Question answering in temporal knowledge graphs requires retrieval that is both time-consistent and efficient. Existing RAG methods are largely semantic and typically neglect expli…