CIVA: Critic-Induced Value-Subspace Attacks on Visual World-Model Agents
Visual world-model agents such as DreamerV3 act through a recurrent latent state rather than a single observation, which weakens frame-wise observation attacks and makes their perturbations vary sharply over time under a strict per-frame perturbation constraint. We study white-box, causal, online attacks on such agents and propose Critic-Induced Value-Subspace Attacks (\textbf{CIVA}). Our key observation is that, along a rollout, critic-guided perturbations concentrate in a low-dimensional subspace induced by the victim's own critic. Based on this observation, CIVA first probes the frozen vict
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzySimilar title/name (fuzzy) · 87%NirDiamant/GenAI_Agents →
“Fuzzy title match (0.94): “CIVA: Critic-Induced Value-Subspace Attacks on Visual World-” ≈ “NirDiamant/GenAI_Agents””
- FuzzySimilar title/name (fuzzy) · 84%Unity-Technologies/ml-agents →
“Fuzzy title match (0.92): “CIVA: Critic-Induced Value-Subspace Attacks on Visual World-” ≈ “Unity-Technologies/ml-agents””
- FuzzyOverlapping authors or contributors · 62%ray-project/ray →
“Shared author/contributor keys: wang”
- FuzzyOverlapping authors or contributors · 62%bytedance/deer-flow →
“Shared author/contributor keys: wang”
- FuzzyOverlapping authors or contributors · 62%mlflow/mlflow →
“Shared author/contributor keys: ruan”
- LinkedLinked via arxiv author · 85%Jiancheng Wang →
“CIVA: Critic-Induced Value-Subspace Attacks on Visual World-Model Agents”
- LinkedLinked via arxiv author · 85%Mingli Zhu →
“CIVA: Critic-Induced Value-Subspace Attacks on Visual World-Model Agents”
- LinkedLinked via arxiv author · 85%Shangtong Zhang →
“CIVA: Critic-Induced Value-Subspace Attacks on Visual World-Model Agents”