8トークンでの計画:潜在世界モデルのためのコンパクトな離散トーカナイザー
Planning in 8 Tokens: A Compact Discrete Tokenizer for Latent World Model
観測をわずか8トークンに圧縮する離散トーカナイザーを提案し、世界モデルを用いた計画の計算コストを大幅に削減して実時間制御を可能にした。
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: Dongwon Kim, Gawon Seo, Jinsung Lee, Minsu Cho, Suha Kwak
分類: cs.CV, cs.AI, cs.RO
原文アブストラクト
World models provide a powerful framework for simulating environment dynamics conditioned on actions or instructions, enabling downstream tasks such as action planning or policy learning. Recent approaches leverage world models as learned simulators, but its application to decision-time planning remains computationally prohibitive for real-time control. A key bottleneck lies in latent representations: conventional tokenizers encode each observation into hundreds of tokens, making planning both slow and resource-intensive. To address this, we propose CompACT, a discrete tokenizer that compresses each observation into as few as 8 tokens, drastically reducing computational cost while preserving essential information for planning. An action-conditioned world model that occupies CompACT tokenizer achieves competitive planning performance with orders-of-magnitude faster planning, offering a practical step toward real-world deployment of world models.
