AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Representing 3D Shapes with 64 Latent Vectors for 3D Diffusion Models

In Cho, Youngbeom Yoo, Subin Jeon, Seon Joo Kim

ICCV 2025 · Image generation › Autoregressive and latent generation · approach: Hierarchical VQ-VAE generation · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use patch-based progressive encoder instead of VecSet cross-attention encoder (ablation)COD-VAE · ShapeNet reconstruction ablation (autoencoder M=32)IoU (%): 91.2 -> 96.1Improved
add more cross-attention layers to the VecSet encoder (ablation)COD-VAE · ShapeNet reconstruction ablation (autoencoder M=32)IoU (%): 93.4 -> 93.5No clear effect
use triplane decoder instead of cross-attention decoder (ablation)COD-VAE · ShapeNet reconstruction ablation (autoencoder M=32)Recon. throughput (sample/s): 3.43 -> 42.68Improved
use 3 cross-attention layers in decoder (ablation)COD-VAE · ShapeNet reconstruction ablation (autoencoder M=32)IoU (%): 96.1 -> 96.3Mixed
increase uncertainty-guided token pruning ratio to 90% (ablation)COD-VAE · ShapeNet reconstruction ablation (autoencoder M=32)IoU (%): 96.2 -> 95.9No clear effect
increase pruning ratio to 75% (R=128) (ablation)COD-VAE · ShapeNet reconstruction ablation (autoencoder M=32)Training speed (iter/s): 4.21 -> 5.60Improved
add two-stage training (ablation)COD-VAE · ShapeNet reconstruction ablation (autoencoder M=32)IoU (%): 94.4 -> 96.1Improved
use reconstruction-based uncertainty instead of confidence-based (ablation)COD-VAE · ShapeNet reconstruction ablation (autoencoder M=32)IoU (%): 95.3 -> 96.1Improved
use learnable latent positions instead of input-dependent (ablation)COD-VAE · ShapeNet reconstruction ablation (autoencoder M=32)IoU (%): 96.0 -> 96.1No clear effect
increase number of latent vectors from 32 to 128 (ablation)COD-VAE · ShapeNet reconstruction ablation (autoencoder M=32)IoU (%): 96.1 -> 96.7Improved
COD-VAE with 64 latents versus VecSet with 64 latentsVecSet · ShapeNet reconstruction (autoencoder)IoU (%): 91.2 -> 96.5Improved
COD-VAE M=64 versus VecSet M=1024 on ObjaverseVecSet · Objaverse reconstruction (VAE)IoU (%): 79.8 -> 79.9Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy