LESSWRONG
LW

Vassil Tashev

Message

Independent researcher | AI Safety Camp 2024, Team 22

Paper review: “The Unreasonable Effectiveness of Easy Training Data for Hard Tasks”

TL;DR: Scalable oversight seems easier based on experiments outlined in a recent paper; questions arise about the implications of these findings. The following graciously provided feedback and advice on the draft, for which I am deeply grateful (in alphabetical order): Sawyer Bernath, Samuel R. Bowman, Bogdan-Ionut Cirstea, Severin Field, Peter...

Feb 29, 2024•11

Vassil Tashev

Independent researcher | AI Safety Camp 2024, Team 22

Vassil Tashev — LessWrong

Vassil Tashev

Message

Independent researcher | AI Safety Camp 2024, Team 22

Paper review: “The Unreasonable Effectiveness of Easy Training Data for Hard Tasks”

Feb 29, 2024•11

Vassil Tashev

Independent researcher | AI Safety Camp 2024, Team 22

Paper review: “The Unreasonable Effectiveness of Easy Training Data for Hard Tasks”

Vassil Tashev

TL;DR: Scalable oversight seems easier based on experiments outlined in a recent paper; questions arise about the implications of these findings.

The following graciously provided feedback and advice on the draft, for which I am deeply grateful (in alphabetical order): Sawyer Bernath, Samuel R. Bowman, Bogdan-Ionut Cirstea, Severin Field, Peter Hase, and Alfie Lamerton.

Introduction

In “The Unreasonable Effectiveness of Easy Training Data for Hard Tasks”, Hase et al., (2024) “study the problem of easy-to-hard generalization, which is relevant for determining how challenging the scalable oversight problem is in the first place, [and] present the surprising conclusion that current language models often generalize relatively well from easy to hard data, even performing as well as 'oracle' models... (read 1179 more words →)