AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Cooperative Training of Fast Thinking Initializer and Slow Thinking Solver for Conditional Learning

Jianwen Xie, Zilong Zheng, Xiaolin Fang, Song Chun Zhu, Ying Nian Wu

TPAMI 2021 · Image generation › Autoregressive and latent generation · approach: Generative latent space tooling · 22 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
add solver Langevin refinement on top of initializer output (ablation)CCoopNets · CIFAR-10Inception score: 6.63 -> 7.30Improved
use more Langevin refinement steps (l=1 or 8 vs 16+) (ablation)CCoopNets · fashion-MNISTtraining stabilityImproved
increase latent dimension d of the initializer (ablation)CCoopNets · fashion-MNISTFIDImproved
increase Langevin steps l from 1 to 64 (d=8) (ablation)CCoopNets · fashion-MNISTtime per epoch (s): 8.98 -> 86.93Hurt
increase latent dimension d from 8 to 128 (l=16) (ablation)CCoopNets · fashion-MNISTtime per epoch (s): 26.88 -> 27.76No clear effect
use a larger Langevin step size (ablation)CCoopNets · fashion-MNISTlearning speedImproved
replace naive concatenation initializer with U-Net with dropout noise (ablation)CCoopNets initializer · image-to-image translationnoise usageImproved
feed ground truth solutions and add l1 loss to the initializer update (ablation)CCoopNets · CMP Facadevisual qualityImproved
CCoopNets vs pix2pix training costCCoopNets · CMP Facadetime per epoch (s): 30.9 -> 32.7Hurt
CCoopNets vs pix2pixpix2pix · CMP Facades inpaintingPSNR: 19.34 -> 20.47Improved
CCoopNets vs pix2pixpix2pix · Paris streetview inpaintingPSNR: 15.17 -> 21.17Improved
CCoopNets vs pix2pix in human preferencepix2pix · CMP Facadepreference ratio: 0.280 -> 0.720Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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