AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Response Attack: Exploiting Contextual Priming to Jailbreak Large Language Models

Miao Ziqi, Lijun Li, Yuan Xiong, Zhenhua Liu, Pengyu Zhu, Jing Shao

AAAI 2026 · Large language models › Language, safety and interpretability · approach: Jailbreak elicitation and risk · 15 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add injected mildly harmful response and trigger prompt (Rharm) (ablation)RA-DRI · HarmBench (avg over 8 LLMs)ASR (%): 48.5 -> 78.8Improved
add prompt rewriting of query into Pinit (ablation)RA-DRI · HarmBench (avg over 8 LLMs)ASR (%): 57.2 -> 78.8Improved
add prompt rewriting of query into Pinit (ablation)RA-SRI · HarmBench (avg over 8 LLMs)ASR (%): 29.9 -> 73.3Improved
add prompt rewriting on GPT-4.1 (ablation)RA-SRI · HarmBench (GPT-4.1)ASR (%): 4.5 -> 71.0Improved
include initial user query Pinit in priming dialogue (ablation)RA-DRI · HarmBench (avg over 8 LLMs)ASR (%): 75.6 -> 78.7No clear effect
generate response from rewritten prompt rather than original query (ablation)RA-DRI · HarmBench (avg over 8 LLMs)ASR (%): 60.6 -> 75.6Improved
use a semantically rich injected response instead of affirmative prefix (ablation)RA-DRI · HarmBench (avg over 8 LLMs)ASR (%): 40.5 -> 78.7Improved
multi-turn chat formatting vs single-turn flattened formats (ablation)RA-DRI · HarmBench (avg over 8 LLMs)ASR (%): 78.7 -> 74.6/74.8No clear effect
RA-DRI vs strongest baseline ActorAttackRA-DRI · HarmBench (avg over 8 LLMs)ASR (%): 82.8 -> 94.8Improved
RA-SRI vs FlipAttack on LLaMA-3-8BRA-SRI · HarmBench (LLaMA-3-8B)ASR (%): 0.0 -> 76.0Improved
RA-SRI vs FlipAttack on GPT-4.1RA-SRI · HarmBench (GPT-4.1)ASR (%): 89.5 -> 88.0No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy