Search papers, labs, and topics across Lattice.
Beijing Institute of Technology
2
0
4
OpenRTAG reveals that traditional GNNs struggle significantly more than LLM-GNNs under realistic data quality degradation, highlighting critical vulnerabilities in graph learning models.
MAGE-RAG achieves a remarkable 52.75% accuracy on LongDocURL by intelligently balancing evidence coverage and context noise in long-document multimodal QA.
