PulseAugur
中
实时 07:00:40
English(EN) DrivingBench: Can Vision-Language Models Drive a Toyota Corolla?

AI模型在新DrivingBench基准测试中接受真实驾驶测试

作者 PulseAugur 编辑部 · [1 个来源] ·

一个名为DrivingBench的新基准已被开发出来，用于测试视觉语言模型在真实世界驾驶的能力。该基准使用一辆丰田卡罗拉，配备了允许模型查看摄像头画面并直接控制车辆转向和速度的工具。在初步测试中，GPT-6 Astra是GPT-6 Astra、Claude Fable 5.1、GPT-5.6 Sol和Grok 4.6中唯一成功完成课程的模型，这表明了当前AI在执行需要持续行动和监控的复杂真实世界任务方面的局限性。 AI

影响 测试视觉语言模型在数字基准之外的真实世界适用性，突显了其在复杂、长时任务方面的当前局限性。

排序理由 该项目描述了一个用于评估AI模型的新基准和研究论文。[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 →

AI模型在新DrivingBench基准测试中接受真实驾驶测试

本文如何被排名

Signal score
25 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该项目描述了一个用于评估AI模型的新基准和研究论文。[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, product, 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) · Aditya Ramabadran, Simon Mahns, Tobias Gessler ·

    DrivingBench：视觉语言模型能开丰田卡罗拉吗？

    arXiv:2609.38948v1 Announce Type: cross Abstract: Frontier models excel at many digital benchmarks, yet their ability to drive a real car, an everyday human skill, remains largely untested. We present DrivingBench, to our knowledge the first benchmark where general-purpose vision…

相关实体

相关话题