[memo]WHEN AN LLM IS APPREHENSIVE ABOUT ITS ANSWERS - AND WHEN ITS UNCERTAINTY IS JUSTIFIED

Abstract

LLMの確信度を判断することは重要です

不確実性のあるタイプに集中す

What estimates, specifically token-wise entropy and model as judge would work for mult

Use LLMs

なんか問題の難しさを定量化してあげよう的なアプローチのようだ

Introduction

Conclusion