Injecting domain knowledge into prompts can substitute for many in-context examples in LLM-based synthetic tabular data generation, cutting required example counts by 40-90%.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
The Prompt is Mightier than the Example
Injecting domain knowledge into prompts can substitute for many in-context examples in LLM-based synthetic tabular data generation, cutting required example counts by 40-90%.