4 papers
Topological Neural Dynamics: A Neuron-wise Framework for Sequence Modeling
Borui Cai, Yao Zhao
Existing sequence models, including RNNs, LSTMs, continuous-time networks, and Transformers, share a common structural principle: layer-wise dynamics, where all neurons in the same…
Beyond Triplet Plausibility: Relation Set Completion in Knowledge Graphs
Zihao Zheng, Borui Cai, Yao Zhao +2
Knowledge graphs (KGs) organize real-world knowledge as triplets and underpin many downstream applications. Due to their inherent incompleteness, knowledge graph completion (KGC) i…
DREAM: LLM-based Dynamic Role-playing via Event-Aware Memory Graph
Zhihao Xiao, Mengting Li, Xintao Wang +4
Role-playing agents (RPAs) have emerged as a key application of large language models, enabling immersive and high-fidelity character simulation. Accurate role-playing of establish…
Intelligence Foundation Model: A New Perspective to Approach Artificial General Intelligence
Borui Cai, Yao Zhao
We propose a new perspective for approaching artificial general intelligence (AGI) through an intelligence foundation model (IFM). Unlike existing foundation models (FMs), which sp…