AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

TrojanTO: Action-Level Backdoor Attacks Against Trajectory Optimization Models

Yang Dai, Oubo Ma, Xingxing Liang, Longfei Zhang, Xiaochun Cao, Shouling Ji, Jiaheng Zhang, Jincai Huang, Li Shen

ICLR 2026 · Adversarial robustness and security › Backdoor and poisoning · approach: Backdoors in large multimodal models · 2 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 TrojanTO post-training attack instead of BaffleTO models · D4RL avg over 6 envs, 3 target actions, 3 TO modelsCP: 0.342 -> 0.701Improved
use TrojanTO instead of IMCTO models · D4RL avg over 6 envs, 3 target actions, 3 TO modelsCP: 0.551 -> 0.701Improved
add trajectory filtering (ablation)TrojanTO · D4RL avgBTP: 0.850 -> 0.914Improved
add batch poisoning (single poisoned transition per batch) (ablation)TrojanTO · D4RL avgASR: 0.528 -> 0.719Improved
add alternating trigger/model training (ablation)TrojanTO · D4RL avgASR: 0.507 -> 0.719Improved
add trajectory filtering (ablation)TrojanTO · D4RL avgASR: 0.678 -> 0.719Mixed
use interior target action '0' instead of boundary '1' (ablation)backdoor · WalkASR: 0.993 -> 0.110Hurt
use trigger dims (1,10,14) instead of (1,2,3) (ablation)backdoor · HalfASR: 0.915 -> 0.000Hurt
use learnable trigger values instead of fixed/dataset triggers (ablation)backdoor · HalfASR: 0.000 -> 0.557Improved
manipulate rewards of target action during backdoor training (ablation)TO model backdoor · WalkASR and BTPNo clear effect
increase trigger noise to 10% (ablation)TrojanTO · WalkASR: 0.980 -> 0.777Hurt
extend persistence of target action to 15 steps (ablation)TrojanTO · WalkCP: 0.993 -> 0.973No 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