#artificial intelligence
Jun 2025
Training-free LLM Verification via Recycling Few-shot Examples
This work proposes a novel framework that Recycles Few-shot examples to verify LLM outputs (ReFeri), which combines a forward confidence score with a backward reconstruction penalty to select candidates that follow few-shot guidance while avoiding demonstration-specific overfitting.
Dongseo Lee, Jimyung Hong, Dongyoung Kim et al.
· arXiv.org · 6 citations