AI Paper Atlas

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

Align to Misalign: Automatic LLM Jailbreak with Meta-Optimized LLM Judges

Hamin Koo, Minseon Kim, Jaehyung Kim

ICLR 2026 · Large language models › Language, safety and interpretability · approach: Jailbreak attacks and defenses · 3 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 the outer-loop scoring template optimization (ablation)AMIS · AdvBench, Claude-3.5-HaikuASR: 86.0 -> 88.0Improved
use the dense 1-10 scoring template instead of binary ASR template (ablation)AMIS · AdvBench, Claude-3.5-HaikuASR: 74.0 -> 88.0Improved
optimize templates at dataset level rather than per query (ablation)AMIS · AdvBench, Claude-3.5-HaikuASR: 84.0 -> 88.0Improved
add prompt inheritance across outer iterations (ablation)AMIS · AdvBench, Claude-3.5-HaikuASR: 80.0 -> 88.0Improved
use a more safety-aligned attacker model (GPT-4o-mini or Claude-3.5-Haiku instead of Llama-3.1-8B) (ablation)AMIS · AdvBench, Claude-3.5-HaikuASR: 88.0 -> 64.0 (GPT-4o-mini), 8.0 (Claude-3.5-Haiku)Hurt
widen inner score range to 1-100 (ablation)AMIS · AdvBench, Claude-3.5-HaikuASR: 88.0 -> 72.0Hurt
narrow inner score range to 1-5 (ablation)AMIS · AdvBench, Claude-3.5-HaikuASR: 88.0 -> 82.0Hurt
use outer-loop normalization constant 10 instead of 100 (ablation)AMIS · AdvBench, Claude-3.5-HaikuASR: 88.0 -> 80.0Hurt
align on both ASR=1 and ASR=0 examples rather than one group (ablation)AMIS · AdvBench, Claude-3.5-HaikuASR: 80.0 / 86.0 -> 88.0Improved
add outer loop on top of a stronger inner loop (K=20, M=10) (ablation)AMIS · AdvBench, Claude-3.5-HaikuASR: 92.0 -> 96.0Improved
add AMIS outer loop to ActorBreaker inner loop (ablation)ActorBreaker · AdvBench, Claude-3.5-Haiku, 1-10 rangeASR: 54.0 -> 58.0Improved
use Llama-3.1-8B as scoring model instead of GPT-4o-mini (ablation)AMIS · AdvBench, Claude-3.5-HaikuASR: 88.0 -> 72.0Hurt

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