PulseAugur
中
实时 14:28:55
English(EN) MMVistaReason: Toward Open-Data and Post-Training Recipes for Multimodal Reasoning

新的基准和框架推动多模态AI代理能力发展

新研究引入了用于评估和改进多模态AI代理在复杂任务中表现的基准和框架。RobotWorld在物理世界交互方面测试代理,揭示了数字能力向机器人技术迁移的差距,其中Astra和Opus 5.5展现出不同的优势。其他研究侧重于通过选择性地迁移训练更新、用于决策规则的代理搜索以及用于多模态图模型的联邦学习来增强视觉推理能力。此外,Hob-VL和VISTA等新基准旨在评估和改进视觉基础推理和交互式世界性能。 AI

影响 这些在多模态代理基准测试和推理框架方面的进展有望加速开发更强大、更可靠的数字和物理任务AI系统。

排序理由 多篇研究论文介绍了用于多模态AI代理和视觉推理的新基准、框架和方法。

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 26 个来源。 我们如何撰写摘要 →

新的基准和框架推动多模态AI代理能力发展

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
多篇研究论文介绍了用于多模态AI代理和视觉推理的新基准、框架和方法。
Source corroboration
26 independent sources
Strong cross-source corroboration — multiple independent publishers covered this within the clustering window.
Topics
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
55 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.
Coverage growth since scoring
+4 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 [26]

  1. arXiv cs.AI TIER_1 English(EN) · Vincenzo Guarino, Emanuele Musumeci, Vincenzo Suriani, Daniele Nardi ·

    iAm.md:通过Agentic内省对未知开放词汇域进行机器人技能自我评估

    arXiv:2610.10962v1 Announce Type: cross Abstract: Agentic AI based on Large Language Model generalization capabilities offers a wide range of potential applications, including planning for embodied tasks. For example, embodied agents based on Foundation models can generate plausi…

  2. arXiv cs.LG TIER_1 English(EN) · Zhiqin Yang, Chenxin Li, Xiaomeng Hu, Yibin Liu, Weidong Huang, Jiankai Sun, Haitao Li, Zijian Wu, Yuzhi Huang, Fanding Huang, Hanwen Sun, Jiashun Liu, Jingqi Tong, Mingxin Huang, Shaoli Hu, Shijue Huang, Tianyi Bai, Xinyuan Wang, Yunlong Lin, Zhengyang … ·

    RobotWorld:跨越不同任务和载体的多模态机器人代理基准测试

    arXiv:2610.10409v1 Announce Type: cross Abstract: General-purpose agents increasingly write code, use tools, and complete complex digital tasks, raising the question of how far these capabilities carry into the physical world. To investigate this, we introduce RobotWorld, a chall…

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

    SuperNav:适用于任何场景下任何任务的代理导航系统

    General-purpose service robots need navigation systems that can handle diverse human requests in unfamiliar environments, combining task generality with scene generality. Some existing methods fine-tune multimodal large language models (MLLMs) to predict navigation actions, makin…

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

    RobotWorld:跨越不同任务和载体的多模态机器人代理基准测试

    General-purpose agents increasingly write code, use tools, and complete complex digital tasks, raising the question of how far these capabilities carry into the physical world. To investigate this, we introduce RobotWorld, a challenging simulation testbed for robot use: turning i…

  5. arXiv cs.AI TIER_1 English(EN) · Suxin Ji, Hungtao Wan, Mingjun Liu, An Zhang ·

    选择性迁移强化学习更新用于视觉推理

    arXiv:2610.08659v1 Announce Type: cross Abstract: Model merging provides a training-free way to transfer reasoning capabilities from language models to vision-language models (VLMs), but endpoint-based transfer can conflate pre-existing model differences with changes acquired dur…

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

    RobotWorld:跨越不同任务和载体的多模态机器人代理基准测试

    General-purpose agents increasingly write code, use tools, and complete complex digital tasks, raising the question of how far these capabilities carry into the physical world. To investigate this, we introduce RobotWorld, a challenging simulation testbed for robot use: turning i…

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

    CueRator:用于适应冻结的多模态编码器的符号规则的代理搜索

    Large language model agents have been used to search over symbolic structures such as programs and equations. We propose CueRator, an agentic framework for policy-aware decision-rule discovery, which adapts frozen contrastive multimodal encoders by searching for the decision rule…

  8. arXiv cs.LG TIER_1 English(EN) · Zekai Chen, Xun Wu, Hailin Zhang, Xunkai Li, Yu Liu, Kairui Yang, Muyan Huang, Xuaner Chen, Rong-Hua Li, Guoren Wang ·

    联邦多模态图基础模型中的感知与推理耦合

    arXiv:2610.00277v1 Announce Type: new Abstract: Federated multimodal graph foundation models (GFMs) aim to adapt pretrained multimodal models to decentralized graph data, where each client owns a private multimodal graph and cannot share raw information. These models typically co…

  9. arXiv cs.AI TIER_1 English(EN) · Yingcheng Liu, Tianyi Jiang, Yujuan Ding, jiangbo Ai, Xun Jiang, Guoqing Wang, Wei Ye, Yi Bin ·

    MoLE:用于互补视觉推理的潜在专家混合模型

    arXiv:2610.01917v1 Announce Type: cross Abstract: Latent visual reasoning equips vision--language models with continuous intermediate states that can process visual evidence without explicit textual reasoning traces or repeated image operations. However, existing methods often al…

  10. arXiv cs.AI TIER_1 English(EN) · Qiushi Han, Keya Hu, Linlu Qiu, Cathy Wu, Kaiming He ·

    VISTA: 在交互式世界中进行推理的视觉工具

    arXiv:2610.02200v1 Announce Type: new Abstract: We show that multimodal models possess strong reasoning abilities and that an appropriate harness can unlock their potential to solve tasks across diverse interactive environments. We introduce VISTA, a visual harness that gives a g…

  11. arXiv cs.AI TIER_1 English(EN) · Yuzhou Wang, Emile Anand, Ijay Narang ·

    Hob-VL:视觉基础布尔推理的基准测试

    arXiv:2610.01605v1 Announce Type: cross Abstract: Reliable visual reasoning requires composing multiple visual observations and returning consistent answers to logically equivalent questions. We introduce Hob-VL, a benchmark for visually grounded Boolean reasoning. Hob-VL compris…

  12. arXiv cs.AI TIER_1 English(EN) · Feiyu Gavin Zhu, Xiaoyu Zhu, Jiqi Yang, Rui Yang, Arnab Kumar Mondal, Yancheng Wang, Xinke Deng, Jean Oh, Reid Simmons, Joerg Liebelt, Xiang Kong, Zhongyu Jiang ·

    基于选择的结构化推理:迈向高效的多模态搜索代理

    arXiv:2610.01892v1 Announce Type: cross Abstract: Multimodal agents commonly generate free-form reasoning before each action. For small models, limited model capacity can result in lengthy reasoning that provides little useful guidance for action generation while incurring substa…

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

    MMVistaReason:迈向多模态推理的开放数据集和训练后方法

    Open multimodal reasoning models have benefited from large-scale reasoning supervision, yet reliable post-training remains challenging due to uneven data quality, inefficient supervision construction, imbalanced difficulty, and cross-domain interference. We introduce MMVistaReaso…

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

    ReCast:面向固定接口的多模态推理的契约保护

    Remote multimodal models offer strong numerical reasoning capabilities over charts and speech, but sending private inputs risks exposing sensitive content. Text-only sanitization cannot directly satisfy fixed media interfaces, while identity anonymization leaves the underlying ta…

  15. arXiv cs.AI TIER_1 English(EN) · Zhihan Zhang, Lizi Liao ·

    通过令牌级感知基础优势估计来增强多模态推理

    arXiv:2609.39168v1 Announce Type: new Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has improved the reasoning capabilities of Multimodal Large Language Models (MLLMs), yet existing frameworks rely on coarse, sequence-level reward signals that lack the fine-grai…

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

    基于选择的结构化推理:迈向高效多模态搜索代理

    Multimodal agents commonly generate free-form reasoning before each action. For small models, limited model capacity can result in lengthy reasoning that provides little useful guidance for action generation while incurring substantial inference cost. To address this challenge, w…

  17. arXiv cs.AI TIER_1 English(EN) · Tianhang Guo, Yulin He, Wei Chen, Wenjuan Zhou, Yuhang Li, Xinbiao Gan ·

    跳过空谈,重新聚焦愿景:多模态大语言模型中用于推理分割的潜在推理

    arXiv:2609.30783v1 Announce Type: cross Abstract: Reasoning segmentation aims to interpret implicit textual queries and enable fine-grained visual perception, which is critical for applications such as human-computer interaction and embodied agents. Existing methods typically gen…

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

    Scaffolding Minds: Optimizing Latent Visual Target Representations for Multimodal Reasoning

    Latent reasoning has advanced multimodal reasoning through a two-stage training paradigm: (1) a helper image is encoded into latent tokens to teach visual chain-of-thought during a supervised fine-tuning (SFT) stage, and (2) these latent tokens are further refined with reward fee…

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

    重新思考潜在视觉推理:将潜在推理与视觉证据联系起来

    Latent visual reasoning (LVR) enables multimodal large language models (MLLMs) to perform intermediate computation in continuous latent tokens rather than expressing every reasoning step in words. However, unlike textual CoT, latent reasoning is not directly observable, making it…

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

    WM-VLM:探究内部世界模型以进行交错的视觉-文本推理

    Humans often solve spatial problems by mentally simulating visual transformations. In contrast, conventional vision-language models (VLMs) reason primarily through language. We investigate whether VLMs can solve spatial problems by reasoning with both text and generated visual st…

  21. arXiv cs.CV TIER_1 English(EN) · Jinkai Zhang, Jingyi Xu, Yuanhong Yu, Jiarui Guo, Ruizhen Hu, Hujun Bao, Xiaowei Zhou, Sida Peng ·

    SuperNav:适用于任何场景下任何任务的代理导航系统

    arXiv:2610.12126v1 Announce Type: cross Abstract: General-purpose service robots need navigation systems that can handle diverse human requests in unfamiliar environments, combining task generality with scene generality. Some existing methods fine-tune multimodal large language m…

  22. arXiv cs.CV TIER_1 English(EN) · Sunchan Park, Beomkwon Cho, Kyeongbo Kong ·

    CueRator:用于适应冻结的多模态编码器的符号规则的代理搜索

    arXiv:2610.07868v1 Announce Type: new Abstract: Large language model agents have been used to search over symbolic structures such as programs and equations. We propose CueRator, an agentic framework for policy-aware decision-rule discovery, which adapts frozen contrastive multim…

  23. arXiv cs.CV TIER_1 English(EN) · Juekai Lin, Honglin Lin, Yuqian Yuan, Xiaolong Wu, Jie Cao, Liang Liang, Yunqi Cao, Yun Zhu, Wenqiao Zhang, Lijun Wu ·

    MMVistaReason:迈向多模态推理的开放数据集和训练后方法

    arXiv:2610.01352v1 Announce Type: new Abstract: Open multimodal reasoning models have benefited from large-scale reasoning supervision, yet reliable post-training remains challenging due to uneven data quality, inefficient supervision construction, imbalanced difficulty, and cros…

  24. arXiv cs.CV TIER_1 English(EN) · Cilin Yan, Yilun Qiu, Wanyang Zhang, Rui Zu, Xiaolong Jiang, Jiayin Cai, Yao Hu ·

    重访分段:工作记忆蒸馏用于推理分段

    arXiv:2609.34863v2 Announce Type: replace Abstract: Multimodal large language models (MLLMs) have approached image segmentation by reasoning about visual content and predicting target locations. Their generated responses contain reasoning traces and localization proposals that ca…

  25. Anyscale blog TIER_1 English(EN) ·

    使用 Anyscale 上的 Ray 扩展机器人决策科学

    Scale robot experiments with Ray on Anyscale, combining parallel simulation and batch inference to reveal what drives policy decisions in minutes.

  26. dev.to — LLM tag TIER_1 English(EN) · trillioniar s ·

    OpenAI的Jalapeno ASIC与实时多模态推理时代

    <p>Today marks a pivotal shift in AI hardware and safety. From the arrival of specialized silicon to the first international regulatory framework for humanoids, the industry is moving from software-centric scaling to integrated physical intelligence.</p> <h2> OpenAI's Jalapeno AS…