Search papers, labs, and topics across Lattice.
Federal University of Campina Grande - UFCG
2
0
5
26
Few-shot prompting boosts LLM performance in generating microservice architectures, achieving an impressive F1 score of 0.97 for service identification.
Structural evaluations of LLM-generated microservices reveal that apparent differences in prompting strategies may mask underlying methodological biases rather than true architectural quality.