相关实验视频
Updated: Jan 31, 2026

05:04
Determining the Mechanical Strength of Ultra-Fine-Grained Metals
Published on: November 22, 2021
2.6K
在视觉和语言导航中,细粒度对齐监督很重要.
IEEE transactions on pattern analysis and machine intelligence
|January 29, 2026
概括
本研究介绍了Landmark-RxR,这是一个用于视觉和语言导航 (VLN) 的细粒度数据集. 细粒度数据显著提高了代理人在VLN任务中的交叉模式对齐能力.
科学领域:
- 计算机视觉 计算机视觉
- 人工智能的人工智能
- 机器人技术 机器人技术 机器人技术
背景情况:
- 视觉和语言导航 (VLN) 要求代理人使用自然语言指令导航3D室内环境.
- 在VLN中,一个关键的挑战是实现代理的轨迹和给出的指令之间的精确的交叉模式对齐.
- 目前的方法由于粗的培训数据,往往受到弱监督.
研究的目的:
- 为了应对在VLN的弱交叉模式对齐监督的挑战.
- 介绍一个新的,人类注释,细粒度数据集VLN.
- 探索和优化训练组件使用细粒度数据,以提高VLN性能.
主要方法:
- 推出了Landmark-RxR,一个细粒度的VLN数据集,带有人类注释,用于精确的监督.
- 研究和调整了核心培训组件:数据增强,培训范式,奖励塑造和导航损失设计.
- 开发了一种针对细粒度VLN数据的新型评估机制.
主要成果:
- 证明Landmark-RxR的细粒度数据有效地提高了代理商的跨模式对齐能力.
- 展示了细粒度监督在提高VLN任务性能方面的优势.
- 验证了拟议的培训策略和评估机制的有效性.
结论:
- 精细的数据对于提高视觉和语言导航中的交叉模式对齐至关重要.
- 标志-RxR数据集为推进VLN研究提供了有价值的细粒度监督.
- 优化训练策略利用细粒度数据,使代理导航能力得到显著改善.
相关概念视频
Classifying Matter by State
103.2K
Chemistry is the study of matter and the changes it undergoes. Matter is anything that has mass and occupies space. Matter is all around us; the air, water, soil, mountains, even our bodies are all examples of matter. Matter is divided into three states — solid, liquid, and gas — that are commonly found on earth. The fourth state of matter, plasma, occurs naturally in the interiors of stars.
103.2K
Language
909
Language is a unique communication system that uses words and systematic rules to organize and transmit information. Unlike other forms of communication, which may involve postures, movements, odors, or vocalizations, language relies on symbols and grammar. This makes human communication distinct from that of other species, who also communicate but do not use language in the same way humans do.
Corballis and Suddendorf (2007) and Tomasello and Rakoczy (2003) highlight the role of language in...
Corballis and Suddendorf (2007) and Tomasello and Rakoczy (2003) highlight the role of language in...
909
Vision
60.0K
Vision is the result of light being detected and transduced into neural signals by the retina of the eye. This information is then further analyzed and interpreted by the brain. First, light enters the front of the eye and is focused by the cornea and lens onto the retina—a thin sheet of neural tissue lining the back of the eye. Because of refraction through the convex lens of the eye, images are projected onto the retina upside-down and reversed.
60.0K
Classifying Matter by Composition
90.3K
Matter: Pure Substances and Mixtures
According to its composition, the matter can be classified into two broad categories — pure substances and mixtures.
A pure substance is a form of matter that has a constant composition throughout with uniform properties. For example, any sample of sucrose has the same composition and same physical properties, such as melting point, color, and sweetness, regardless of the source from which it is isolated.
A mixture is composed of two or...
According to its composition, the matter can be classified into two broad categories — pure substances and mixtures.
A pure substance is a form of matter that has a constant composition throughout with uniform properties. For example, any sample of sucrose has the same composition and same physical properties, such as melting point, color, and sweetness, regardless of the source from which it is isolated.
A mixture is composed of two or...
90.3K
Physical and Chemical Properties of Matter
166.3K
The characteristics that enable us to distinguish one substance from another are called properties.
166.3K
The Atomic Theory of Matter
127.9K
The earliest recorded discussion of the basic structure of matter comes from ancient Greek philosophers. Leucippus and Democritus argued that all matter was composed of small, finite particles that they called atomos, meaning “indivisible.” Later, Aristotle and others came to the conclusion that matter consisted of various combinations of the four “elements” — fire, earth, air, and water — and could be infinitely divided. Interestingly, these philosophers...
127.9K

