ETA-VLA: 時間融合とLLM内スパース化による視覚言語行動モデルの効率的トークン適応
ETA-VLA: Efficient Token Adaptation via Temporal Fusion and Intra-LLM Sparsification for Vision-Language-Action Models
自動運転向けVLAモデルの計算負荷を減らすため、過去フレームの冗長な視覚トークンをテキストガイドと時間的一貫性で間引くフレームワークを提案し、精度を保ちつつ推論FLOPsを大幅削減した。
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: Yiru Wang, Anqing Jiang, Shuo Wang, Yuwen Heng, Zichong Gu, Hao Sun
分類: cs.RO, cs.AI
原文アブストラクト
The integration of Vision-Language-Action (VLA) models into autonomous driving systems offers a unified framework for interpreting complex scenes and executing control commands. However, the necessity to incorporate historical multi-view frames for accurate temporal reasoning imposes a severe computational burden, primarily driven by the quadratic complexity of self-attention mechanisms in Large Language Models (LLMs). To alleviate this bottleneck, we propose ETA-VLA, an Efficient Token Adaptation framework for VLA models. ETA-VLA processes the past $n$ frames of multi-view images and introduces a novel Intra-LLM Sparse Aggregator (ILSA). Drawing inspiration from human driver attention allocation, ILSA dynamically identifies and prunes redundant visual tokens guided by textual queries and temporal consistency. Specifically, we utilize a text-guided scoring mechanism alongside a diversity-preserving sparsification strategy to select a sparse subset of critical tokens, ensuring comprehensive awareness of the driving scene. Extensive experiments on the NAVSIM v2 demonstrate that ETA-VLA achieves driving performance comparable to state-of-the-art baselines while reducing computational FLOPs by approximately 32\%. Notably, our method prunes 85% of visual tokens and reduces inference FLOPs by 61\%, but still retaining 94% of the original accuracy on the NAVSIM v2 benchmark.
関連論文
- 言語の役割を再考する:自動運転のための効率的VLAへ向けてVLA/自動運転
- VLA運転のためのマルチ軌道教師信号とポリシー最適化の整合VLA/自動運転
