AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

Beware Untrusted Simulators -- Reward-Free Backdoor Attacks in Reinforcement Learning

Ethan Rathbun, Wo Wei Lin, Alina Oprea, Christopher Amato

ICLR 2026 · Adversarial robustness and security › Backdoor and poisoning · approach: Backdoors in large multimodal models · 5 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 Daze dazing instead of reward poisoning (TrojDRL-C)Daze · Hopper (continuous, PPO)attack success rate: 25.6% -> 94.1%Improved
use Daze instead of SleeperNets-CPPO agent · Half Cheetahattack success rate: 3.3% -> 92.4%Improved
use Daze instead of SleeperNets-C on IntersectionPPO agent · Intersectionattack success rate: 99.8% -> 92.3%Hurt
use SleeperNets-C instead of Daze (benign return cost)PPO agent · Intersectionbenign return: 238.4 -> 158.6Hurt
add Daze to Atari agents vs no poisoningPPO agent · Pacmanbenign return: 628.7 -> 591.8No clear effect
add Daze vs TrojDRL on Q*bertPPO agent · Q*bertattack success rate: 88.2% -> 99.3%Improved
use Daze vs Q-Incept on Q*bertPPO agent · Q*bertattack success rate: 100.0% -> 99.3%Hurt
increase daze factor k from 1 to 16 (ablation)Daze · Hopper (beta=1.0%)attack success rate: 40.29% -> 95.56%Improved
increase poisoning rate beta from 1.0% to 2.0% at k=16 (ablation)Daze · Hopperattack success rate: 95.56% -> 85.82%Hurt
increase poisoning rate beta from 1.0% to 2.0% at k=16 (ablation)Daze · Hopperbenign return: 2222.4 -> 1407.7Hurt
switch victim algorithm from PPO to TD3Daze · Hopperattack success rate: 94.1% -> 99.9%Improved
switch victim algorithm from PPO to DQNDaze · Breakout (beta=1%)attack success rate: 97.6% -> 90.1%Mixed

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