跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.24240 · v1

ViLFlow: Parallel Vision-Language Generation with Continuous Latent Flows

首次提交 2026-11-18 19:19
版本更新 2026-11-18 19:19
雷达观测 2026-11-19 15:04
arXiv ↗PDF ↗

摘要

Vision-language generation has predominantly relied on autoregressive and discrete diffusion models. Autoregressive models generate tokens sequentially, while discrete diffusion models typically predict tokens independently within each denoising step. We propose ViLFlow (Vision-Language Flow), a continuous-flow vision-language model that combines joint modeling of output embeddings with parallel token decoding. ViLFlow adapts pretrained continuous language generation to visual question answering and image captioning through learned image guidance and prediction-aware token decoding. A visually adapted teacher provides image-guided predictions as additional denoising supervision. The denoiser learns to reproduce these predictions directly, retaining one image-conditioned prediction per generation step. We train the token decoder on denoiser predictions and paired reference text to improve the conversion of continuous outputs into words. The decoder can be adapted independently, improving answering and captioning without further denoiser training. For captioning, it also supplies text feedback during late denoising, allowing decoded words to inform subsequent refinement. ViLFlow outperforms autoregressive and masked-diffusion baselines by up to 8.46 percentage points on VQAv2 and 4.50 points on GQA without GQA training. On COCO Karpathy test, it improves CIDEr by 16.25 over masked diffusion and 17.81 over sampled autoregression.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-18 19:19 · 最新观测版本

ViLFlow: Parallel Vision-Language Generation with Continuous Latent Flows

Vision-language generation has predominantly relied on autoregressive and discrete diffusion models. Autoregressive models generate tokens sequentially, while discrete diffusion models typically predict tokens independently within each denoising step. We propose ViLFlow (Vision-Language Flow), a continuous-flow vision-language model that combines joint modeling of output embeddings with parallel token decoding. ViLFlow adapts pretrained continuous language generation to visual question answering and image captioning through learned image guidance and prediction-aware token decoding. A visually adapted teacher provides image-guided predictions as additional denoising supervision. The denoiser learns to reproduce these predictions directly, retaining one image-conditioned prediction per generation step. We train the token decoder on denoiser predictions and paired reference text to improve the conversion of continuous outputs into words. The decoder can be adapted independently, improving answering and captioning without further denoiser training. For captioning, it also supplies text feedback during late denoising, allowing decoded words to inform subsequent refinement. ViLFlow outperforms autoregressive and masked-diffusion baselines by up to 8.46 percentage points on VQAv2 and 4.50 points on GQA without GQA training. On COCO Karpathy test, it improves CIDEr by 16.25 over masked diffusion and 17.81 over sampled autoregression.

查看此版本解读 →查看此版本原文 ↗