EquiVLA: 回転同変ビジョン・言語・行動モデルのための汎用フレームワーク
EquiVLA: A General Framework for Rotationally Equivariant Vision-Language-Action Models
凍結した視覚言語バックボーンとフローマッチング拡散トランスフォーマー行動ヘッドを組み合わせた任意のアーキテクチャに適用可能な、エンドツーエンドのSO(2)同変VLAモデルの最初の汎用フレームワークを提案。回転に対する幾何学的帰納バイアスを導入し、データ効率と汎化性能を大幅に向上させた。
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: Thien-Loc Ha, Quang-Tan Nguyen, Trong-Bao Ho, Long Dinh, Minh Duc Nguyen, Gia-Binh Nguyen, Pham Tri Quang, Minh N. Vu, Duy M. H. Nguyen, An Thai Le, Ngo Anh Vien
分類: cs.RO
原文アブストラクト
Vision-Language-Action (VLA) models have emerged as a powerful paradigm for generalist robot manipulation, yet they lack geometric inductive biases: policies trained at specific orientations require substantially more data to generalize across rotational configurations. We present \textsc{EquiVLA}, the first general framework for end-to-end $\mathrm{SO}(2)$-equivariant VLA models, applicable to any architecture coupling a frozen vision-language backbone with a flow-matching Diffusion Transformer action head. \textsc{EquiVLA} introduces \textsc{EquiPerceptor}, which produces approximately $\mathrm{SO}(2)$-equivariant visual representations from frozen ViT features; and \textsc{EquiActor}, an exactly $\mathrm{SO}(2)$-equivariant flow-matching Diffusion Transformer action head. Together, they establish an approximate $\mathrm{SO}(2)$ equivariance chain from camera observations to predicted action sequences. Instantiated on GR00T~N1.5 and evaluated across four LIBERO suites, CALVIN ABCD$\to$D, and five real-robot tasks on Mobile ALOHA, \textsc{EquiVLA} achieves $92.6\%$ average success on LIBERO (vs. $78.1\%$ baseline), an average sequence length of $4.03$ on CALVIN (vs. $3.45$), and improves real-robot success from $54\%$ to $72\%$.
