ZipTok3D：具有緊湊標記前綴的高保真度3D標記化

摘要

緊湊的token序列對於高效的3D生成至關重要。然而，現有的3D tokenizer通常將潛在表示組織在空間區域上，或組織為固定大小的全域token集合，但這兩者在壓縮到極低的token預算時，重建品質都會急遽下降。在本論文中，我們提出了ZipTok3D，這是一種專門設計用於從極短token序列中進行高保真重建的3D tokenizer。其核心思想是將物體幾何結構組織為資訊量漸增的全域token前綴，並透過迭代解碼展開這些緊湊表示。具體而言，巢狀丟棄（nested dropout）在訓練期間於編碼後隨機截斷潛在序列，並要求每個保留的前綴都能重建完整物體，因而優先將必要的幾何資訊置於前端的token中。接著，解碼器重複套用一個參數共享的Transformer區塊，從每個前綴還原出細粒度的幾何細節，而不需要額外的生成式採樣階段。在相同的token維度下，ZipTok3D在ShapeNet上僅用一個token、在TRELLIS上僅用四個token，便能達到與32-token的COD-VAE基線相當的重建品質，分別產生短32倍與8倍的token序列。

English

Compact token sequences are essential for efficient 3D generation. However, existing 3D tokenizers typically organize latent representations either over spatial regions or as fixed-size sets of global tokens, both suffering sharp reconstruction degradation when compressed to extremely low token budgets. In this paper, we present ZipTok3D, a 3D tokenizer designed for high-fidelity reconstruction from extremely short token sequences. Its key idea is to organize object geometry into progressively informative global-token prefixes and unfold these compact representations through iterative decoding. Specifically, nested dropout randomly truncates the latent sequence after encoding during training and requires each retained prefix to reconstruct the complete object, thereby prioritizing essential geometric information in the leading tokens. The decoder then repeatedly applies a parameter-shared Transformer block to recover fine-grained geometry from each prefix without a separate generative sampling stage. With the same token dimension, ZipTok3D achieves reconstruction quality comparable to the 32-token COD-VAE baseline using only one token on ShapeNet and four on TRELLIS, yielding 32times and 8times shorter token sequences, respectively.