Jianwei Tai

The paper establishes a theoretical information-theoretic bound proving that for Vision-Language-Action (VLA) models, capability and robustness cannot both be arbitrarily high, quantifying the trade-off between performance and adversarial resilience.

The paper identifies a 'deployment-safety gap' in Vision-Language-Action (VLA) policies, showing that identical model checkpoints can result in physically different and unsafe robot actions due to action unnormalization and controller conventions.

The paper establishes an information-theoretic upper bound on the combined functional capacity and perturbation retention of code LLMs, quantifying the security budget available for code generation.

The paper investigates the fragility and recovery mechanisms of chain-of-thought (CoT) answer hijacking, demonstrating that specific problem cells are susceptible to targeted recovery and that source control significantly influences the hijacking mechanism.