両刃の剣としての推論:視覚-言語-行動モデルにおけるアーキテクチャと段階横断的ロバスト性
Reasoning as a Double-Edged Sword: Architecture and Cross-Stage Robustness in Vision-Language-Action Models
推論ステップを追加するとVLAモデルのロバスト性が向上するかを、3つのモデル(推論なし、テキストCoT、潜在反復ループ)で比較し、潜在反復モデルが最も脆弱であることを示した。また、推論出力を安全信号として利用する防御手法の限界も明らかにした。
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: Tuan Duong Trinh, Naveed Akhtar, Basim Azam
分類: cs.RO, cs.AI, cs.CR, cs.LG
原文アブストラクト
Does adding a reasoning step make a Vision-Language-Action (VLA) model more robust to perturbation? Intuitively, a policy that reasons before acting should absorb a perturbed input better than one that maps observations directly to actions. We test this premise head-on across three models that span the reasoning spectrum (no reasoning, a text chain-of-thought, and a latent iterative loop), perturbing each at the vision, reasoning, and action stages on LIBERO and SimplerEnv. Two questions organize the study: does the reasoning design shift robustness, and can the reasoning be read back at runtime as a safety signal? We find that the latent-iterative model is by far the least robust: under both stochastic noise and white-box perturbation its task success collapses, while the other two hold. This fragility is structural rather than cumulative: varying the reasoning depth at inference barely moves it. Reasoning outputs can in principle be monitored, but the monitors fail under fair tests. A plan--action consistency probe that looks near-perfect under naive evaluation falls to chance under adaptive attack. Under matched-FPR calibration, fusing it with an action-anomaly probe never lifts defended success above undefended. Scoped to these output-level behavioral probes under white-box vision-stage attack, this ceiling is a precondition that any viable defense must first satisfy.
