English(EN) Beyond Post-Hoc Rewards: Privileged Reasoning Exploration for Multimodal Embeddings

新的PILOT框架通过引导式推理探索增强多模态嵌入

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-26 04:00

研究人员开发了PILOT，一个旨在通过改进生成多模态嵌入的推理过程来增强多模态嵌入的新框架。与依赖事后奖励的先前方法不同，PILOT将推理建模为潜在变量，利用仅输入和输入加关系证据来指导探索。该方法在MMEB-V2基准测试上进行了测试，在各种任务中显示出显著的改进，特别是在视觉-文档检索方面，证明了在训练过程中进行关系监督的价值。 AI

影响这项研究可能带来更具信息量的多模态嵌入，提高视觉-文档检索等任务的性能，并可能应用于其他跨模态应用。

排序理由该集群包含一篇详细介绍新框架和基准结果的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 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: 13 / 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新框架和基准结果的学术论文。[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: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Tianyu Yang, Yijun Tian, Lars Benedikt Kaesberg, Jan Philip Wahle, Bela Gipp, Terry Ruas, Daniel Kurzawe · 2026-10-26 04:00

超越事后奖励：多模态嵌入的特权推理探索

arXiv:2610.32474v1 Announce Type: new Abstract: Advances in Multimodal Large Language Models (MLLMs) have driven substantial progress in multimodal embeddings. Recent methods exploit generative capabilities by augmenting embedded inputs with self-generated reasoning to produce mo…

本文如何被排名

报道来源 [1]

超越事后奖励：多模态嵌入的特权推理探索

相关实体

相关话题