English(EN) Where It Breaks: Diagnosing Perception, Memory, and Decision Failures of a Vision-Language Agent in Pac-Man

Qwen3.5-9B 智能体在吃豆人游戏中的感知和决策失败得到诊断

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-05 05:00

研究人员开发了一种诊断视觉语言智能体失败的方法，使用了一个经过训练以玩吃豆人游戏的 Qwen3.5-9B 模型。该研究区分了感知错误（读取屏幕）和决策错误（选择动作）。通过分离这些组件，研究团队发现，虽然该智能体能够准确读取静态游戏屏幕，但在游戏过程中仍然做出了糟糕的动作选择。这凸显了需要更细致的训练后评估，将感知与策略性能分开，并考虑完整的游戏验证，而不是单一的选定分数。 AI

影响提供了一个识别和解决人工智能智能体特定故障模式的框架，可能改进训练和评估过程。

排序理由学术论文，详细介绍了一种用于人工智能智能体的新诊断方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 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: 26 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 学术论文，详细介绍了一种用于人工智能智能体的新诊断方法。[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: 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]

arXiv cs.LG TIER_1 English(EN) · Xinglu Wang, Edward Wang, Yong Zhang, Zhenan Fan · 2026-11-05 05:00

症结何在：诊断视觉语言智能体在吃豆人游戏中的感知、记忆与决策失误

arXiv:2611.06704v1 Announce Type: new Abstract: Training an agent to play Pac-Man with only up, down, left and right actions raises a basic question: when performance is poor, what should be improved? Difficulty can arise from reading the screen or from choosing actions over a lo…

本文如何被排名

报道来源 [1]

症结何在：诊断视觉语言智能体在吃豆人游戏中的感知、记忆与决策失误

相关实体

相关话题