新的 DexHoldem 基准评估具身AI在真实扑克任务中的表现

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-01 04:00

研究人员推出 DexHoldem，一个旨在评估具身AI系统在真实世界灵巧操作任务中表现的新基准，特别是在德州扑克场景下。该基准包含一套全面的演示，共计1,470个，涵盖14种操作原语，一个标准化的物理策略基准，以及一个代理感知基准。初步结果显示，$\pi_{0.5}$策略在原语执行任务完成率方面达到最高，为61.2%，而Opus 5.5在代理感知准确性方面以微弱优势领先。 AI

影响该基准有望加速开发更强大的具身AI代理，以应对复杂、真实的现实世界任务。

排序理由该集群描述了一个新的AI研究学术基准。[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 →

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; 该集群描述了一个新的AI研究学术基准。[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]

arXiv cs.AI TIER_1 English(EN) · Feng Chen, Tianzhe Chu, Li Sun, Pei Zhou, Zhuxiu Xu, Shenghua Gao, Yuexiang Zhai, Yanchao Yang, Yi Ma · 2026-10-01 04:00

DexHoldem：用于德州扑克灵巧操作的代理机器人基准测试

arXiv:2605.18727v2 Announce Type: replace-cross Abstract: Evaluating embodied systems with real dexterous hardware requires more than isolated motor-skill tests: an agent must perceive a changing scene (e.g. a tabletop), choose a context-appropriate action, execute it with a dext…

本文如何被排名

报道来源 [1]

DexHoldem：用于德州扑克灵巧操作的代理机器人基准测试

相关实体

相关话题