Xiao Ma, Gao Huang, Zanlin Ni, Zeyu Liu, Cheng Zhong, Xin Deng, Yeguo Hua
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. If it links a repository, it is listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Discrete visual tokenizers transform images into a sequence of tokens, enabling token-based visual generation akin to language models. However, this process is inherently challenging, as it requires both compressing visual signals into a compact representation and discretizing them into a fixed set of codes. Traditional discrete tokenizers typically learn the two tasks jointly, often leading to unstable training, low codebook utilization, and limited reconstruction quality. In this paper, we introduce CODA (COntinuous-to-Discrete Adaptation), a framework that decouples compression and discretization. Instead of training discrete tokenizers from scratch, CODA adapts off-the-shelf continuous VAEs-already optimized for perceptual compression-into discrete tokenizers via a carefully designed discretization process. By primarily focusing on discretization, CODA ensures stable and efficient training while retaining the strong visual fidelity of continuous VAEs. Empirically, with 6× less training budget than standard VQGAN, our approach achieves a remarkable codebook utilization of 100% and notable reconstruction FID (rFID) of 0.43 and 1.34 for 8× and 16× compression on ImageNet 256× 256 benchmark.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2503.17760")
get_code_for_paper("2503.17760")
have("2503.17760")
Connect an agent — have() is free.