VLA-FAIL: 微調整された視覚言語行動モデルのための効率的なタスク失敗検出
VLA-FAIL: Efficient Task Failure Detection for Finetuned Vision-Language-Action Models
視覚言語行動モデル（VLA）の実行時失敗を、失敗データを必要とせず軽量に検出するフレームワークを提案。最終層のマハラノビス距離と行動チャンクの一貫性という2つの検出器を組み合わせる。
詳しい要約
1. どんなもの？
2. 先行研究と比べてどこがすごい？
3. 技術・手法の肝は？
4. どうやって有効だと検証した？
5. 議論はある？
6. 次に読むべき論文は？
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Florian Seligmann, Emiliyan Gospodinov, Enes Ulas Dincer, Gerhard Neumann
分類: cs.LG, cs.CV
原文アブストラクト
Vision-language-action models (VLAs) achieve state-of-the-art performance on many robotic manipulation tasks, yet they can still behave unpredictably in out-of-distribution scenarios. Runtime failure detection is therefore essential for the safe real-world deployment of VLAs. However, existing task failure detectors require computationally expensive action sampling, are based on architectural assumptions that limit their applicability to VLAs, or need access to failure rollouts. We propose VLA-FAIL, a lightweight and broadly applicable failure detection framework for VLAs that combines two novel failure detectors with minimal overhead, without requiring failure data. The first, last-layer Mahalanobis distance (LLMD), detects out-of-distribution states by measuring token-wise deviations in last-layer features relative to the training data. The second, action chunk consistency (ACC), exploits the temporal overlap induced by receding-horizon control and detects failures when consecutive action chunks become inconsistent. To capture the trade-off between detection accuracy and detection latency, we introduce AUCPDT, a threshold-independent metric that jointly evaluates precision, recall, and detection time. Through extensive real-world and simulation experiments, we demonstrate that LLMD and ACC capture complementary failure modes whose combination enables reliable and early failure detection across diverse tasks, frequently outperforming significantly more expensive baseline methods.
