BadWAM: When World-Action Models Dream Right but Act Wrong
World-action models (WAMs) are emerging as a promising foundation for embodied control: rather than predicting actions alone, they learn representations that couple action generation with future world prediction. This coupling is often viewed as a source of robustness, interpretability, and safety, as a robot's action can in principle be checked against its imagined future. In this paper, we show that this assumption is fragile. We introduce BadWAM, a unified framework for modeling and evaluating World-Action Drift Attacks: a new class of WAM-specific adversarial attacks that use small visual
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.
- FuzzyOverlapping authors or contributors · 62%bytedance/deer-flow →
“Shared author/contributor keys: wang”
- FuzzyOverlapping authors or contributors · 62%ray-project/ray →
“Shared author/contributor keys: wang”
- LinkedLinked via arxiv author · 85%Yongqi Li →
“BadWAM: When World-Action Models Dream Right but Act Wrong”
- LinkedLinked via arxiv author · 85%Xingyi Yang →
“BadWAM: When World-Action Models Dream Right but Act Wrong”
- LinkedLinked via arxiv author · 85%Xinchao Wang →
“BadWAM: When World-Action Models Dream Right but Act Wrong”
- FuzzySimilar title/name (fuzzy) · 84%liguodongiot/llm-action →
“Fuzzy title match (0.92): “BadWAM: When World-Action Models Dream Right but Act Wrong” ≈ “liguodongiot/llm-action””