Weixi Zhang

Beyond Retrieval: Evidence Access and Event Aggregation in Agent Memory

When Agent Memory Needs to Compute: Evidence Access and Executable Aggregation

Questions about an agent's history often require aggregating events, which retrieval alone does not resolve. We combine source-preserving evidence access with executable aggregation. Trajectory Memory (TM) retains the original trajectory and constructs a source-addressable reading view at question time. TallyMem augmen...

Revisiting Degree-Corrected Spectral Clustering: a Condition-Free Spectral Analysis and Extension

This study explores an alternative condition-free analysis for the clustering quality of DCSC from a pure spectral view, without any random graph models, and proposes ASCENT (Adaptive Spectral Clustering with Node-wise correcTion), a simple yet effective extension of DCSC.

FlatLand: Personalized Graph Federated Learning via Tailored Lorentz Space

This work proposes FlatLand, a novel personalized federated learning method that embeds different clients'data in tailored Lorentz space of hyperbolic geometry, and develops a parameter decoupling strategy that separates heterogeneous information from common knowledge, enabling direct aggregation without requiring clie...