Search papers, labs, and topics across Lattice.
4
0
10
Heterogeneous graph neural networks can get a boost in attribute completion by strategically sampling and reweighting node types based on their information content.
AI agents that ace isolated coding tasks fall apart when faced with the messy reality of continuous software evolution, dropping from 80% to 38% success rates in a new benchmark.
Achieve up to 68x faster and 170x more energy-efficient SAR image recognition on FPGAs without sacrificing adversarial robustness, thanks to a new model-hardware co-design framework.
Forget independent learners: Action Graph Policies unlock dramatically better multi-agent coordination by explicitly modeling action dependencies.