Search papers, labs, and topics across Lattice.
3
0
4
Results show that structured schema context and lightweight prompting can substantially reduce reliance on fine-tuning for scalable conversational access to KGs, though closing the remaining gap to fully fine-tuned approaches will likely require reducing dependence on curated question-query exemplars.
Extracting historical person-place relations reveals that lightweight classifiers can outperform large language models in specific contexts, challenging assumptions about model size and performance.
Can your relation extraction system handle noisy, multilingual historical texts while remaining accurate and computationally efficient?
