Wei Zhao
5 papers · Latest:
MinT: Managed Infrastructure for Training and Serving Millions of LLMs
MinT is a managed infrastructure system for efficiently training and serving millions of LoRA-adapted LLMs over shared base models.
Empowering Heterogeneous Graph Foundation Models via Decoupled Relation Alignment
DRSA enhances heterogeneous graph foundation models by decoupling feature semantics from relation structures, preventing "Type Collapse" and "Relation Confusion."
ClawGuard: A Runtime Security Framework for Tool-Augmented LLM Agents Against Indirect Prompt Injection
ClawGuard is a runtime security framework protecting tool-augmented LLM agents from indirect prompt injection by enforcing rules at tool-call boundaries.
Are Independently Estimated View Uncertainties Comparable? Unified Routing for Trusted Multi-View Classification
TMUR introduces unified routing to address comparability issues in independently estimated view uncertainties for trusted multi-view classification.
Plasticity-Enhanced Multi-Agent Mixture of Experts for Dynamic Objective Adaptation in UAVs-Assisted Emergency Communication Networks
PE-MAMoE uses a plasticity-enhanced multi-agent mixture of experts to improve UAV communication network adaptation to dynamic user demands.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.