AI Paper Atlas

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

A Game-Theoretic Analysis of Attacks on Large Language Models via Compositional Skills

Xinbo Wu, Huan Zhang, Abhishek Umrawal, Lav R. Varshney

ICML 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Adversarial prompts for language models

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
use GPT-4.1 instead of Llama-3-70B as the helpfulness rater (ablation)LLM-as-rater pipeline · JailbreakBench-based rater datasetagreement with human experts: 47% -> 89%Improved
use best-response (skill-probing) attack instead of PAIR, GCG, JBC, PRS in the 1-skill setupadversarial prompting · JBB-Behaviors, target LLMsBin-JR scoreImproved
expand the skill space and mix two skills per intent (ablation)best-response attack · JBB-Behaviorsacceptance rate, Bin-JR, JR scoreImproved
add the misleading-the-attacker defense (ablation)prompt and response filtering defender · all evaluated target LLMsattack performance reductionImproved
compare with iterative optimization attack PAIR on JR scorebest-response attack · VicunaJR scoreHurt
increase defender capacity cdefender (theory) · equilibrium value without skill transferattacker gain: J* = 1 - c/|S(k)| max p(i)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