MineExplorer: Meituan LongCat Evaluation Reveals Capability Gaps of Top Multimodal Large Models in Dynamic Open Worlds
Original title:让AI离开温室,走向动态世界:MineExplorer揭示顶级多模态大模型被忽视的能力断层
AI summary
The Meituan LongCat team built MineExplorer, the first evaluation benchmark for minute-level long-horizon tasks in open worlds, including 813 manually verified instances (1-hop to 4-hop) and a rule-based milestone automatic evaluation framework.
Key Takeaways
- The Meituan LongCat team built MineExplorer
- the first evaluation benchmark for minute-level long-horizon tasks in open worlds
- including 813 manually verified instances (1-hop to 4-hop) and a rule-based milestone automatic…
Quick Facts
- Published
- Updated
- Source
- 公众号:龙猫LongCat(美团)
- Category
- Papers & Research
- AI score
- 62
感知强于推理、导航失败占比近60%,把具身智能的瓶颈清晰地拉回到规划能力,而非感官或资源。
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: MineExplorer: Meituan LongCat Evaluation Reveals Capability Gaps of Top Multimodal Large Models in Dynamic Open Worlds. July 23, 2026. https://new.nat6.net/en/news/cmsdu2t2a011frofz9mrkr93n/