Search papers, labs, and topics across Lattice.
School of Information and Electrical Engineering, Hebei University of Engineering, Handan, China
4
0
11
15
Traditional evaluation methods may misrepresent LLM capabilities, but RepBench reveals a more nuanced understanding through a multi-benchmark approach that uncovers 182 capability clusters.
LGRID transforms the landscape of local-life service recommendations by generating interpretable Semantic IDs that are both semantically rich and collision-resistant.
ClawRec achieves a remarkable 11.3% increase in recommendation quality over traditional systems by leveraging cross-platform user behavior for more relevant suggestions.
A novel AIGC service provision scheme reduces average task completion latency in Cyber-Physical Manufacturing by integrating advanced reinforcement learning techniques.
