AI Paper Atlas

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

LLM-Agnostic Semantic Representation Attack

Jiawei Lian, Jianhong Pan, Lefan Wang, Yi Wang, Tairan Huang, Shaohui Mei, Lap‐Pui Chau

TPAMI 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Adversarial prompts for language models · 0 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
use Vicuna 7B as a semantic proxy model (SRA+ vs SRA)SRA · HarmBench, Llama 2 7BASR (%): 30.33 -> 100.00Improved
use Vicuna 7B as a semantic proxy model (SRA+ vs SRA)SRA · HarmBench, Llama 2 13BASR (%): 19.00 -> 100.00Improved
use Vicuna 7B as a semantic proxy model (SRA+ vs SRA)SRA · HarmBench, Llama 2 70BASR (%): 17.33 -> 100.00Improved
use Vicuna 7B as a semantic proxy model (SRA+ vs SRA)SRA · HarmBench, Qwen3 4B ThinkingASR (%): 66.33 -> 100.00Improved
use Vicuna 7B as a semantic proxy model (SRA+ vs SRA)SRA · HarmBench, Llama 3.1 8BASR (%): 45.00 -> 100.00Improved
use Vicuna 7B as a semantic proxy model (SRA+ vs SRA)SRA · HarmBench, R2D2 7BASR (%): 42.00 -> 93.00Improved
use SRA instead of GCGGCG · HarmBench, Llama 2 7BASR (%): 46.25 -> 30.33Hurt
replace GCG with SRA under 15s budgetGCG · AdvBench, Vicuna 7Bperplexity: 753.39 -> 24.21Improved
replace BEAST with SRA under 15s budgetBEAST · AdvBench, Vicuna 13BASR (%): 37.69 -> 86.73Improved
increase budget from 15s to 60sSRA · AdvBench, Vicuna 7Bperplexity: 24.21 -> 18.67Improved
lower semantic threshold tau from 1000 to 10 (ablation)SRA · Vicuna 13B, 15s budgetASR (%): 75.38 -> 86.15Improved
lower semantic threshold tau from 100 to 20 (ablation)SRA · Vicuna 13B, 15s budgetASR (%): 77.69 -> 86.73Improved

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