SOURCE-LINKED INTELLIGENCE

StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training

arXiv · AI, language, vision and robotics · article · Sep 22, 2026 · UTC

Vector Quantization (VQ) is fundamental to discrete visual tokenizers that power modern autoregressive and masked image generation models. While recent shared-projection codebook methods have substantially advanced codebook utilization, training stability remains a critical and underexplored challenge. We argue that the root cause lies in the entanglement of the Encoder--Decoder and Codebook training: because neither module can reliably fulfill its own responsibility in isolation, the system can only function when the two subsystems happen to cooperate---a fragile condition that breaks down pr

Read original source ↗ Open in workspace

recordType

paper

region

Global