TRAP is a tail-aware ranking attack that plants a backdoor in world models so that a trigger causes the model to reorder a few critical imagined trajectories and redirect planning while preserving normal behavior on clean inputs.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
A method using convex autoencoders and kernel-based learning creates a finite abstraction in latent space that overapproximates unknown dynamical systems, enabling scalable verification with correctness guarantees that transfer back to the original system.
citing papers explorer
-
TRAP: Tail-aware Ranking Attack for World-Model Planning
TRAP is a tail-aware ranking attack that plants a backdoor in world models so that a trigger causes the model to reorder a few critical imagined trajectories and redirect planning while preserving normal behavior on clean inputs.
-
Verification of Unknown Dynamical Systems via Autoencoder Latent Space
A method using convex autoencoders and kernel-based learning creates a finite abstraction in latent space that overapproximates unknown dynamical systems, enabling scalable verification with correctness guarantees that transfer back to the original system.