AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

MirrorShield: Towards Dynamic Adaptive Defense Against Jailbreaks via Entropy-Guided Mirror Crafting

Rui Pu, Chaozhuo Li, Rui Ha, Litian Zhang, Lirong Qiu, Xi Zhang

AAAI 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: LLM jailbreak and injection defense · 1 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 length constraint to mirror generation (ablation)MirrorShield · BaitAttack (Llama2)ASR: 0.654 -> 0.009Improved
add syntax constraint to mirror generation (ablation)MirrorShield · BaitAttack (Llama2)ASR: 0.243 -> 0.009Improved
add sentiment constraint to mirror generation (ablation)MirrorShield · BaitAttack (Llama2)ASR: 0.071 -> 0.009Improved
add multiple-query guidance (ablation)MirrorShield · BaitAttack (Vicuna)ASR: 0.042 -> 0.028Improved
use MirrorShield instead of SafeDecodingVicuna-13b-v1.5 · Average ASR over 12 datasets/attacksASR: 0.083 -> 0.036Improved
use MirrorShield on well-aligned modelLlama2-7b-chat · Average ASR over 12 datasets/attacksASR: 0.315 -> 0.006Improved
use MirrorShield instead of no defenseLlama2-7b-chat · AlpacaEvalWinRate: 69.6% -> 68.6%Hurt
use Paraphrase defenseLlama2-7b-chat · AlpacaEvalWinRate: 69.6% -> 34.1%Hurt
raise RIU threshold sigma from 0.80 to 0.90 (ablation)MirrorShield · 100 benign + 100 jailbreak prompts (Llama2)detection accuracy: 0.97 -> 0.75Hurt
increase candidate mirrors to 5 or more (ablation)MirrorShield · BaitAttack (Llama2)ASR: below 0.3 at 4 candidates, near zero at 5+Improved

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