Table 11 Zero-shot vs few-shot performance.

From: A zero-shot LLM framework for multimodal grievance classification, urgency scoring, and abuse detection in civic feedback systems

Setting

Accuracy (%)

MAE

Zero-shot (Proposed)

92.4

0.041

8-shot

77.9

0.093

32-shot

84.5

0.067

128-shot

89.8

0.052