English(EN) Thinking with Images as Continuous Actions: Numerical Visual Chain-of-Thought

新的NV-CoT框架使MLLMs能够通过连续坐标对图像进行推理

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-16 04:00

研究人员推出了一种新颖的数值视觉思维链（NV-CoT）框架，该框架使多模态大语言模型（MLLMs）能够使用连续数值坐标对图像进行区域基础推理。该方法将模型的动作空间从离散的token扩展到连续的欧几里得空间，只需极少的架构更改即可直接生成边界框坐标。NV-CoT支持监督微调和强化学习，利用高斯或拉普拉斯策略以及重参数化采样进行策略优化。实验表明，与现有的视觉推理基线相比，NV-CoT显著提高了定位精度、最终答案准确性和训练收敛性。 AI

影响该框架可以提高多模态AI系统中视觉推理的精度和效率。

排序理由该集群包含一篇详细介绍多模态LLM新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 14 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍多模态LLM新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CV TIER_1 English(EN) · Kesen Zhao, Beier Zhu, Junbao Zhou, Xingyu Zhu, Zhongqi Yue, Hanwang Zhang · 2026-10-16 04:00

将图像视为连续动作进行思考：数值化视觉思维链

arXiv:2602.23959v2 Announce Type: replace Abstract: Recent MLLMs increasingly rely on visual chain-of-thought to perform region-grounded reasoning over images. However, existing approaches ground regions via either textified coordinates (causing modality mismatch and semantic fra…

本文如何被排名

报道来源 [1]

将图像视为连续动作进行思考：数值化视觉思维链

相关实体

相关话题