PulseAugur
中
实时 09:33:22

新的人类视觉感知基准HumanSightBench揭示MLLM在视觉感知方面存在困难

一个新基准HumanSightBench已被开发出来,用于评估多模态大型语言模型(MLLM)的视觉感知能力。该基准包含2,876个视觉问题,这些问题对人类来说很容易,但对模型来说却很有挑战性,并且经过了五个纳入门槛的筛选,以尽量减少歧义和捷径等混淆因素。初步评估显示,即使是表现最好的MLLM在计数等任务上也表现困难,得分低于27.5%,这凸显了与人类感知相比,它们在视觉理解方面存在显著差距。 AI

影响 凸显了MLLM视觉感知方面的显著差距,为未来的训练和评估工作提供了指导。

排序理由 该集群描述了一个用于评估AI模型的新基准,属于研究范畴。[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 →

新的人类视觉感知基准HumanSightBench揭示MLLM在视觉感知方面存在困难

本文如何被排名

Signal score
16 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该集群描述了一个用于评估AI模型的新基准,属于研究范畴。[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, other
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]

  1. arXiv cs.AI TIER_1 English(EN) · Zixin Su, Yangguang Lin, Zegang Peng, Yuntong Sun, Xuanyi Wang, Jiahao Wang, Qiankai Xu, Bing Yuan, Jingjia Huang, Ruyang Liu, Yanxu Hu, Yufeng Zhou, Yi Lin ·

    HumanSightBench:用于多模态大语言模型中人类视觉感知能力的基准测试

    arXiv:2611.12305v2 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) have demonstrated strong performance on knowledge-intensive and reasoning-based vision-language benchmark datasets, but still struggle to solve the visual questions that people can …