Repository Issue Activity (beta)

#183 回来吧 牢田！中国的AI需要你！😭😭😭

#182 🚀 Your VAR Model is Secretly an Efficient and Explainable Generative Classifier — [ICLR 2026]

#181 [Resource] Awesome-Visual-Autoregressive-Modeling — A curated paper list built on VAR

#167 请教高手：我使用作者的vae、vard16_pth但是跑的结果accm只有acc_mean ≈5%，而作者论文中的是acc_mean ≈ 82.6%

#180 Cannot reproduce "without CFG" FID in Table 3 (expected ~4.64, got ~12), while "with CFG" is reproducible

#178 🌟 Introducing SAR (Self-Autoregressive Refinement): stable self-rollout that unlocks robust post-training for Visual Autoregressive image generation.

#179 📢 Introducing SSG (Scaled Spatial Guidance) for Multi-Scale VAR Generation via Information-Theoretic Analysis

#177 🌈 Introducing DiverseVAR, a training-free framework that unleashes the inherent generative diversity of Visual Autoregressive models while preserving fidelity and text–image alignment.

#164 🚀 Introducing scale and spatial Markovian assumptions into VAR — achieving faster generation with reduced memory consumption.

#137 Small Scale Training - POC

#176 xformer and flash_attention are not used for low resolution

#45 FID misalignment

#174 Introducing "Next Visual Granularity Generation"

#132 Some doubts about quantization loss

#175 can you share the multi-scale VQ-VAE training and evaluation code?

#143 VAR training

#59 There was no increase in speed after installing flash-attn and xformer.

#173 🚀 Introducing SCALAR: A Scale-wise Controllable VAR Framework for Next-Scale Prediction

#171 how to modify VAR for image translation task ?

#170 Any plans to release the VQVAE training code?

#28 Training code for VAE

#169 Introducing HMAR: Efficient Hierarchical Masked AutoRegressive Image Generation, that builds on VAR.

#71 Computation and Memory Consumption When Training Models

#166 🚀 Introducing an efficient KV cache compression framework tailerd for next-scale prediction.

#136 加载权重问题