PulseAugur
中
实时 08:52:13
English(EN) OneSearch-VL: Unified Multimodal Deep Research Agent for Image and Video

新的多模态模型和研究推动了人工智能在边缘、推理和生成方面的能力

几篇研究论文和一个新模型发布,专注于推进多模态人工智能能力。Liquid Foundation Models 发布了开源的 d1-3B 模型,该模型针对边缘设备进行了优化,能够处理文本和图像,并具有快速的推理时间。同时,新的研究探索了在嘈杂监督下改进多模态学习(REFINE)、对直观视觉推理进行基准测试(Humanity's Sixth Sense)、全模态嵌入的结构化专业化(Syn-Omni)、通过张量建模学习多模态协同作用(HRIL)以及将视觉世界模型融入 LLM(WOVEN)的方法。此外,MUNITE 提供了一个用于任意到任意多模态生成的统一框架,PrismSSL 提供了一个用于多模态自监督学习的库。 AI

影响 多模态人工智能的进步正在提高推理、生成和效率,从而在各个领域实现更复杂اً的应用。

排序理由 多篇关于多模态人工智能进展的arXiv研究论文。

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 27 个来源。 我们如何撰写摘要 →

新的多模态模型和研究推动了人工智能在边缘、推理和生成方面的能力

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
多篇关于多模态人工智能进展的arXiv研究论文。
Source corroboration
27 independent sources
Strong cross-source corroboration — multiple independent publishers covered this within the clustering window.
Topics
model release, paper, product
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
56 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.
Coverage growth since scoring
+2 source(s) since last score
New sources have picked up this story since our last re-score. Score will update on the next scoring pass.

完整方法见我们的编辑标准。

报道来源 [27]

  1. Hugging Face Blog TIER_1 English(EN) ·

    面向边缘的多模态开放d1决策模型

  2. arXiv cs.CL TIER_1 English(EN) · Zheyu Fan, Yue Zhang, Mingkai Deng, Kangrui Wang, Qineng Wang, Canyu Chen, Jie Hao, Xing Fan, Chenlei Guo, Eric P. Xing, Mohit Bansal, Manling Li ·

    WOVEN:将视觉世界建模编织进多模态大语言模型

    arXiv:2610.12417v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) struggle with spatial, embodied, physical, and temporal reasoning. We hypothesize that these failures reflect a shared deficit in visual transition reasoning, and test whether this capabili…

  3. arXiv cs.AI TIER_1 English(EN) · Xingang Guo, Jing Gu, Brian Jang, Renxiong Wang, Utkarsh Tyagi, Daniel Quigley, Steven Li, David Yan, Daniel Yue Zhang, Darvin Yi, Forrest Huang, HiJae Kim, Tianyi Zhang, Jared Lichtarge, Jihua Huang, Le Xue, Manan Tomar, Qiuyi Richard Zhang, Ruofei Yu, … ·

    人类的第六感:多模态模型直觉视觉推理的基准测试

    arXiv:2610.08966v2 Announce Type: replace Abstract: Humans perceive far more in a scene than what is explicitly depicted: a single glance captures past causes and future trajectories; a quick peek determines if a vehicle can fit between two parked cars; a few seconds of video rev…

  4. arXiv cs.LG TIER_1 English(EN) · Jiashuo Zou, Xiaobo Xia ·

    在嘈杂监督下的多模态学习中学习信任什么

    arXiv:2610.11057v1 Announce Type: cross Abstract: Multimodal classification processes and relates information from multiple modalities to achieve more accurate predictions. However, existing methods typically rely on high-quality ground-truth labels, which are difficult to obtain…

  5. arXiv cs.AI TIER_1 English(EN) · Qun Dai, Liangjian Wen, Jiang Duan, Yong Dai, Dongkai Wang, Maolin Wang, Mingjie Wang, Jianzhuang Liu, He Yan, Zhao Kang ·

    HRIL:通过高阶张量建模学习多模态协同

    arXiv:2610.12393v1 Announce Type: new Abstract: Self-supervised multimodal representation learning has achieved remarkable success across diverse domains, yet capturing synergistic information remains challenging due to the complexity of cross-modal interactions. Unlike the share…

  6. arXiv cs.AI TIER_1 English(EN) · Youngtaek Oh, Qiyu Wu, Hiromi Wakaki, Junmo Kim, Yuki Mitsufuji ·

    Syn-Omni:用于全模态嵌入的结构化专业化和渐进式协作

    arXiv:2610.12256v1 Announce Type: cross Abstract: Omnimodal embeddings naturally involve both shared representations and modality-specific features across heterogeneous inputs. However, existing omnimodal embedding methods often rely on a single shared parameter space over mixed-…

  7. arXiv cs.IR (Information Retrieval) TIER_1 English(EN) · Yuki Mitsufuji ·

    Syn-Omni:用于全模态嵌入的结构化专业化和渐进式协作

    Omnimodal embeddings naturally involve both shared representations and modality-specific features across heterogeneous inputs. However, existing omnimodal embedding methods often rely on a single shared parameter space over mixed-modality data, limiting structural separation betw…

  8. arXiv cs.LG TIER_1 English(EN) · Kyeongmin Yeo, Minhyuk Sung ·

    MUNITE:统一的多模态潜在推理,用于任意到任意的多模态生成

    arXiv:2610.09866v1 Announce Type: new Abstract: We introduce MUNITE, a latent-variable framework for flexible any-to-any multimodal generation that treats encoding and latent generation as the same inference problem under different amounts of observed evidence. Given any subset o…

  9. arXiv cs.LG TIER_1 English(EN) · Melika Shirian, Kianoosh Vadaei, Kian Majlessi, Audrina Ebrahimi, Peyman Adibi, Hossein Karshenas ·

    PrismSSL:一个接口,多种模态;用于多模态自监督学习的单接口库

    arXiv:2511.17776v2 Announce Type: replace Abstract: We present PrismSSL, a Python library that unifies state-of-the-art self-supervised learning (SSL) methods across audio, vision, graphs, and cross-modal settings in a single, modular codebase. The goal of the demo is to show how…

  10. Perplexity blog TIER_1 English(EN) ·

    超越单一向量的多模态嵌入

    A new family of late-interaction models for text and image retrieval, with a shared embedding space across model sizes.

  11. Hugging Face Daily Papers TIER_1 English(EN) ·

    OneSearch-VL:统一的多模态图像和视频深度研究代理

    Single-image, multi-image, and video deep research require different visual operations but share a workflow of visual grounding, external retrieval, and fact composition. A key challenge is to preserve the dependencies linking localized visual anchors, entity relations, source-su…

  12. arXiv cs.LG TIER_1 English(EN) · Payal Mohapatra, Haodong Yang, Yueyuan Sui, Stephen Xia, Benjamin Lundell, Qi Zhu ·

    通过自适应获取和顺序融合实现高效多模态推理

    arXiv:2610.07466v1 Announce Type: new Abstract: Multimodal systems often encode every available input, even when a subset suffices for prediction. Adaptive acquisition can reduce this cost by using predictions from incrementally fused evidence to decide which modality to encode n…

  13. arXiv cs.AI TIER_1 Italiano(IT) · Chufan Shi, Cheng Yang, Tiannuo Yang, Isadora White, Yiwei Chen, Taylor Berg-Kirkpatrick, Xuezhe Ma ·

    统一多模态模型中的视觉弃权

    arXiv:2610.07887v1 Announce Type: cross Abstract: Unified multimodal models (UMMs) integrate understanding and generation, yet their generative behavior is rarely governed by what they understand about the task. We formalize visual abstention: when a requested visual transformati…

  14. arXiv cs.AI TIER_1 English(EN) · Jinghao Xu, Zhenhua Guo, Xiaofeng Zhu, Xiaoshuang Shi ·

    面向多模态学习的动态对齐与校准

    arXiv:2610.07928v1 Announce Type: cross Abstract: Dynamic multimodal learning aims to learn robust representations by adaptively modeling information discrepancies across modalities. However, existing methods still suffer from two limitations: (i) static cross-modal alignment str…

  15. arXiv cs.CL TIER_1 English(EN) · Chenlong Wang, Yuhang Chen, Zhihan Hu, Dongping Chen, Wenhu Chen, Sarah Wiegreffe, Tianyi Zhou ·

    量化统一多模态模型中理解与生成之间的差距

    arXiv:2602.02140v2 Announce Type: replace Abstract: Recent advances in unified multimodal models (UMM) have demonstrated remarkable progress in both understanding and generation tasks. However, whether these two capabilities are genuinely aligned and integrated within a single mo…

  16. Hugging Face Daily Papers TIER_1 English(EN) ·

    VisionWeave:将弹性视觉表示作为多模态大语言模型的原生能力进行编织

    Multimodal large language models have become the dominant paradigm for visual understanding, but incur substantial costs by encoding inputs into dense, fixed-size patch tokens. However, visual information is unevenly distributed: some regions require fine-grained detail, while ot…

  17. arXiv cs.CL TIER_1 English(EN) · Huijuan Wang, Chufan Shi, Cheng Yang, Yaokang Wu, Taylor Berg-Kirkpatrick, Xuezhe Ma ·

    统一多模态模型中的递归自改进

    arXiv:2610.03002v1 Announce Type: new Abstract: Unified multimodal models (UMMs) understand and generate both text and images, which lets a model produce its own training data. Existing self-improvement in UMMs keeps supervision on the visual side, where image understanding judge…

  18. arXiv cs.LG TIER_1 English(EN) · Zhongjing Gu, Fengqiang Wan, Yiming Cui, Yufa Feng, Yang Yang ·

    通过功能性进展平衡多模态学习

    arXiv:2610.03035v1 Announce Type: new Abstract: Multimodal learning often suffers from modality imbalance, where the joint optimization process is dominated by a single modality. Existing methods typically estimate modality imbalance from score disparities derived from prediction…

  19. Hugging Face Daily Papers TIER_1 English(EN) ·

    Salt++:面向少样本流式多模态生成的上下文对齐后训练

    Few-step streaming audio--video generation requires both causal modeling and step distillation, yet standard training recipes face two context-related challenges. Teacher forcing pairs clean history with a noisy target, but supervises predictive contextual representations only in…

  20. Hugging Face Daily Papers TIER_1 English(EN) ·

    使用证据对齐读出学习多模态嵌入

    Multimodal large language models can expose task-relevant evidence through generation, but producing useful evidence does not by itself determine how it enters a retrieval embedding. We study whether the semantic organization of that evidence can also specify where representation…

  21. arXiv cs.CV TIER_1 English(EN) · Yicheng Xue, Han Wu, Jufeng Yang, Minjing Dong, Xinghao Chen, Hanting Chen, Jianyuan Guo ·

    回顾与前瞻:按需视觉记忆助力高效多模态推理

    arXiv:2610.12060v1 Announce Type: new Abstract: Processing long visual token sequences from high-resolution images makes multi-step reasoning computationally expensive for multimodal Large Language Models (MLLMs). Existing one-shot pruning and aggregation methods compress visual …

  22. arXiv cs.CV TIER_1 English(EN) · Hongyu Li, Manyuan Zhang, Kaituo Feng, Shu Chen, Dian Zheng, Hao Li, Hao Yu, Zhangquan Chen, Zoey Guo, Ray Zhang, Shaofei Huang, Tianrui Hui, Linjiang Huang, Si Liu ·

    OneSearch-VL:统一的多模态深度研究图像和视频智能体

    arXiv:2610.12419v1 Announce Type: new Abstract: Single-image, multi-image, and video deep research require different visual operations but share a workflow of visual grounding, external retrieval, and fact composition. A key challenge is to preserve the dependencies linking local…

  23. arXiv cs.CV TIER_1 (CA) · Nikita Kisel, Illia Volkov, Klara Janouskova, Jiri Matas ·

    多模态大语言模型作为图像分类器

    arXiv:2603.06578v3 Announce Type: replace Abstract: Multimodal Large Language Model (MLLM) classification performance depends critically on evaluation protocol and ground truth quality. Studies comparing MLLMs with supervised and Vision-Language Models (VLMs) report conflicting c…

  24. arXiv cs.CV TIER_1 English(EN) · Naim Reza, Md Al Amin, Ho Yub Jung ·

    MCL: Meta Convolution Layer

    arXiv:2610.11117v1 Announce Type: new Abstract: Dynamic convolution enhances convolutional neural networks (CNNs) by adapting kernels to input content, but it expresses the effective kernel as a linear mixture of a small number of basis kernels, which limits expressivity and comp…

  25. arXiv cs.CV TIER_1 English(EN) · Xiangtao Kong, Shuaizheng Liu, Rongyuan Wu, Lingchen Sun, Zhengqiang Zhang, Jinxin Zhao, Yuhui Wu, Lei Zhang ·

    HarnessIR:利用多模态基础模型实现通用真实世界图像修复

    arXiv:2610.10133v1 Announce Type: new Abstract: Real-world low-quality images suffer from complex mixed degradations, including but not limited to noise, blur, atmospheric effects, etc. Recent agentic methods usually model real-world image restoration (Real-IR) as a sequential to…

  26. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    认识 Cloudflare Clef-omni:一个用于音频、视频、图像和文本的 30B-A3B 多模态决策模型

    <p>Cloudflare released Clef-omni, an Apache-2.0 multimodal decision model built on Qwen3-Omni-30B-A3B. It scores text, images, audio and video in one call for $0.15 per million input tokens, while Clef-flash drops to $0.038.</p> <p>The post <a href="https://www.marktechpost.com/2…

  27. dev.to — LLM tag TIER_1 Italiano(IT) · KAMAL KISHOR ·

    Ollama 多模态模型:本地运行视觉 AI

    <p>Dealing with complex tasks that require understanding both text and images usually means hitting external APIs, racking up costs, and dealing with potential data privacy concerns. Building intelligent agents that can interpret visual information alongside natural language, and…