MAE I Trust Myself? Self-Evaluating VLA Action Generation with Markov Attention Entropy

Vision-Language-Action models (VLAs) integrate visual perception, language instruction, and action generation into end-to-end policies across heterogeneous architectures. However, enabling VLAs to self-evaluate their action generation reliability without external supervision remains a major challenge. Existing methods either rely on expert annotations or estimate uncertainty only from output statistics, largely ignoring internal signals. In this work, we observe that internal visual modality entropy exhibits consistent distinctions between successful and failed tasks across heterogeneous VLAs. Although VLAs' architectures differ in their action generation, we show that they share a common latent action generation abstraction evolving under visual perception and language instruction, which we formulate as a Conditional Generative Markov Chain. Based on this formulation, we propose Markov Attention Entropy, a self-evaluation framework that directly converts internal attention signals into architecture-aware reliability scores, and introduce LIBERO-Reflect, a balanced benchmark with positive and negative episodes for evaluating episode-level reliability ranking. Extensive experiments across heterogeneous VLAs architectures and diverse scenarios show that MAE consistently outperforms state-of-the-art baselines on AUPR, AUROC, and FPR@95.