日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

VLAarXiv:2611.18891

言語能力を取り戻す:パラメータマージによるVLAモデルのVLM能力復元

Bring Back Language to Actions: Restoring VLM capabilities in VLA Models via Parameter Merging

シェア:XThreadsFacebookLINEはてブBluesky

VLMとそのVLA派生モデルのパラメータを再学習なしでマージし、タスクベクトルをSVD分解して混合係数を最適化することで、行動予測と視覚言語能力を両立させる手法を提案。

著者: Mohamed Salim Aissi, Olivier Sigaud, Laure Soulier, Nicolas Thome

分類: cs.AI, cs.CV

原文アブストラクト

Vision-Language-Action models (VLAs) are obtained by fine-tuning pre-trained Vision-Language Models (VLMs) on robot demonstration data, enabling action prediction from visual observations and language instructions. However, this fine-tuning induces catastrophic forgetting, substantially degrading the semantic and visual grounding capabilities of the original VLM. We address this by merging the parameters of a VLM and its VLA derivative without retraining either model. Concretely, we extract the task vector (the difference between VLA and VLM weights) and decompose it via Singular Value Decomposition. For each component, we learn a scalar mixing coefficient controlling how much of the VLA adaptation is retained, optimised to jointly preserve action prediction and vision-language capabilities. Experiments show that our merged model performs well on both axes simultaneously, consistently outperforming established merging baselines.

関連論文

  1. 1Boston Dynamics、車輪付きAtlasの研究試作機を紹介研究・技術
  2. 2Google DeepMind、ロボット向けAIを発表新製品・発表
  3. 3GXOとApptronik、Apolloの物流現場試験を拡大 最大1,000台の展開を構想量産・展開
  4. 4パナソニックHDなど、触覚に即応する両手作業AI「T-Rex」を共同開発研究・技術
  5. 5SoftBank主導、Agile Robotsが5億ドル調達 NVIDIAと日立も参加資金調達

PR・アフィリエイト広告

FRONTIER(GPU搭載BTOパソコン)

FRONTIER(GPU搭載BTOパソコン)の公式広告バナーFRONTIERでGPU搭載PCを見る ↗
PR本紙発行元 EmplifAI