視覚予測型VLAモデルのテスト時学習
Test-Time Training for Visual Foresight Vision-Language-Action Models
視覚予測型VLAモデルのOOD脆弱性を、テスト時に予測画像と観測のペアを用いた自己教師あり学習で改善する手法を提案。
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: Sangwu Park, Wonjoong Kim, Yeonjun In, Sein Kim, Hongseok Kang, Chanyoung Park
分類: cs.CV, cs.LG, cs.RO
原文アブストラクト
Visual Foresight VLA (VF-VLA) has become a prominent architectural choice in the recent VLA due to its impressive performance. Nevertheless, the inherent design of VF-VLA makes it particularly vulnerable to out-of-distribution (OOD) shifts. Because the quality of action directly depends on the accuracy of the predicted future visual information, OOD conditions affect both stages at once. To address this vulnerability, we propose Test-Time Training Visual Foresight VLA ($T^3$VF), a test-time training approach motivated by the observation that the predicted future image and its subsequent observation form a natural supervision pair. To further address the practical challenges that arise from indiscriminate test-time updates, we introduce an adaptive update filtering mechanism. Empirically, $T^3$VF mitigates the OOD vulnerability of VF-VLA at a modest additional inference cost, without requiring any architectural modification or auxiliary modules.
