研究人员引入了可变码本大小量化（VCQ）来解决自回归视觉生成模型的局限性。VCQ沿序列动态修改码本大小，显著提高了ImageNet等数据集上的重建性能并降低了gFID分数。此外，VVS和推测性耦合解码（SCD）等新方法通过优化推测性解码技术，在保持生成质量的同时减少所需的前向传播次数，从而加速了这些模型的推理速度。 AI
影响 量化和推测性解码方面的这些进步有望带来更快、更高效的视觉生成模型，可能降低推理成本并实现新的应用。
排序理由 该集群包含多篇arXiv论文，详细介绍了自回归视觉生成和推测性解码技术方面的新研究。
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
arXiv:2605.06207v1 Announce Type: cross Abstract: Most discrete visual tokenizers rely on a default design: every position in the sequence shares the same codebook. Researchers try to scale the codebook size $K$ to get better reconstruction performance. Such a constant-codebook d…
Most discrete visual tokenizers rely on a default design: every position in the sequence shares the same codebook. Researchers try to scale the codebook size $K$ to get better reconstruction performance. Such a constant-codebook design hits a fundamental information-theoretic lim…
arXiv:2511.13587v3 Announce Type: replace Abstract: Visual autoregressive (AR) generation models have demonstrated strong potential for image generation, yet their next-token-prediction paradigm introduces considerable inference latency. Although speculative decoding (SD) has bee…
arXiv:2510.24211v2 Announce Type: replace Abstract: Autoregressive (AR) modeling has recently emerged as a promising new paradigm in visual generation, but its practical adoption is severely constrained by the slow inference speed of per-token generation, which often requires tho…