AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

Rapid Poison: Practical Poisoning Attacks Against the Rapid Response Framework

David Huang, Jaewon Chang, Avidan Shah, Prateek Mittal, Chawin Sitawarin

ICML 2026 · Adversarial robustness and security › Backdoor and poisoning · approach: Backdoors in large multimodal 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
add few-shot MCQ examples to the injection instead of zero-shot description (ablation)Format poisoning · MMLU domainsFPR: 94-100% FPR, ~30-60% above zero-shotImproved
use zero-shot format description up to 5% poisoning (ablation)Format poisoning · MMLU domainsFPR: many domains below 60-70%Hurt
use held-out WildChat queries as poison instead of unrelated synthetic safe queries (ablation)ChatGPT name poisoning · Name-based FPR on ClaudeFPR: 48% -> 19%Improved
use contrastive query augmentation instead of naive few-shot nouns (ablation)ChatGPT name poisoning · Name-based FPR on ClaudeFPR: 70% -> 19%Improved
add LLM-guided concept embedding instead of naive HP name insertion (ablation)Omission attack HP · PAIR jailbreaksFNR: 14.7% -> 55.3%Improved
use paraphrased longer trigger instead of short n-gram trigger (ablation)Omission attack human-like style · Cipher jailbreakFNR: 0% -> 40.67%Improved
use paraphrased trigger Meshy instead of Midjourney (ablation)Omission attack · Harmful queriesFNR: 96% -> 98%No clear effect
add greedy 3-gram set-cover sampling to injection (ablation)Distribution poisoning · low poisoning ratesFPRImproved
use Meta SecAlign 70B as proliferation model (ablation)Domain poisoning · Professional law FPRFPR: 98% -> 0%Improved
raise poisoning rate from 1% in omission (ablation)Direct poisoning vs RR pipeline · Harmful queriesFNR: 62-63% harmful vs 94-100% jailbreaksMixed
poison SWE-Bench coding prompts then test on WildChatDistribution poisoning · WildChatFPR: 100% -> 0%No clear effect

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