RAIN, 로봇 작업 일반화를 위한 영역인식 상호작용 네트워크
RAIN: Robotic Task Generalization with Region-Aware Interaction Networks
- 매체
- arXiv
- 게시
- 중요도
- 2/5
- 화제 지수
- —
한국어 요약
VLA 모델이 훈련 궤적을 암기하는 문제를 해결하기 위해 목표 지정 부작업 시퀀스와 영역 마스크로 작업을 표현하는 RAIN을 제안했다. 새로운 객체와 장면 배치에 학습한 기술을 재사용하고 미학습 작업으로 조합 가능하며, LIBERO-Analogy 벤치마크와 실제 테이블탑 작업에서 기존 VLA보다 강한 일반화를 달성했다.
원문 발췌 · English
Existing vision-language-action (VLA) models benefit from large-scale vision-language pretraining, but tend to memorize training trajectories, limiting their ability to adapt learned interactions to new task and scene configurations. To address this problem, we propose \textbf{Region-Aware Interaction Networks (RAIN)}, which formulates a task as a sequence of goal-conditioned subtasks, each specified by an action type and a target-region mask. This formulation allows learned skills to be reused across different objects and scene layouts and composed into task sequences not seen during training. Its Target-adaptive Cross-View Encoder (TCE) conditions multi-view features on target masks and exchanges information across views, while PlanDiT predicts target-directed trajectories and actions. A Transition Head estimates subtask completion to enable automatic transitions. We also introduce a \textbf{Reference-based Retargeting Strategy (RRS)}, which uses recorded trajectories to synthesize smooth approaches to reference interactions with alternative target regions, without collecting additional data. Experiments on \textbf{LIBERO-Analogy}, our new benchmark for robotic task generalization, and real-world tabletop tasks demonstrate that RAIN generalizes more robustly than existing VLA baselines.