Reward Stealing Attack on Large Language Models

Paper recorded by Signals 4 on 2026-10-05 in cs.CL. Abstract reproduced from arXiv; link to the original below.

Published 2026-10-05 on arXiv · recorded by Signals 4 on 2026-10-06

Category: cs.CL · 自然语言处理 · first seen 2026-10-06

Abstract

Adversarial attacks on Large Language Models (LLMs) aim to induce harmful content. However, existing methods suffer from high computational costs or strict model-pairing dependencies, limiting their scalability and transferability. We propose Reward Stealing Attack (ReSA), an adversarial attack framework that targets the latent safety reward underlying LLM alignment. ReSA employs maximum entropy i

Read on arXiv →