亚马逊云科技创新“神经稀疏检索”：文本匹配实现语义搜索，助推AI发展新时代
近日，亚马逊云科技在AI领域取得了新的突破，推出了一项名为“神经稀疏检索”的创新技术。该技术通过文本匹配就能实现语义搜索，为用户带来更为智能和精准的搜索体验。
随着大模型的火热发展，语义检索逐渐成为一项热门技术。特别是在RAG（retrieval augmented generation）应用中，检索结果的相关性对于AI生成的最终效果具有至关重要的作用。对此，亚马逊云科技在其OpenSearch项目中引领创新，推出神经稀疏检索技术。
传统的语义检索方案大多利用语言模型将文本编码为高维向量，并采用近似k-邻近搜索进行检索。然而，面对高昂的VectorDB和语言模型部署费用，许多用户望而却步。而亚马逊云科技推出的神经稀疏检索技术，能够在降低成本的同时，实现高效的语义搜索。
据了解，该技术作者是来自OpenSearch中国研发团队的机器学习负责人杨扬博士以及机器学习工程师耿志超和管聪。他们凭借深厚的学术背景和丰富的实践经验，成功研发出这一创新技术。
目前，该技术已经在全球范围内得到了广泛关注和应用。超过5亿的下载量以及70个以上的企业合作伙伴，证明了该技术的实力和影响力。
亚马逊云科技表示，未来将继续在AI领域深耕，推出更多创新技术，为用户提供更为优质的服务。同时，也欢迎更多的优秀人才加入他们的团队，共同推动AI技术的发展。
机器之心AIxiv专栏作为全球学术交流与传播的重要平台，将持续关注并报道亚马逊云科技在AI领域的最新进展。如果您有优秀的工作想要分享，欢迎投稿或者联系报道。
相信在亚马逊云科技等企业的努力下，语义检索技术将不断进步，为人们的生活带来更多便利和智能体验。
英语如下：
News Title: “Amazon Cloud Innovation: Neural Sparse Retrieval for Text Matching and Semantic Search Revolution”
Keywords: Neural Sparse Retrieval, Semantic Retrieval Technology, Machine Learning for Text Matching
News Content:
Amazon Web Services (AWS) has achieved a new breakthrough in AI, introducing an innovative technology called “Neural Sparse Retrieval.” Through text matching, this technology enables semantic search, bringing a more intelligent and precise search experience to users.
With the rapid development of large models, semantic retrieval has gradually become a popular technology. Especially in RAG (retrieval augmented generation) applications, the relevance of search results plays a crucial role in the final effect of AI generation. In response, AWS has led innovation in this field through its OpenSearch project and introduced Neural Sparse Retrieval technology.
Traditional semantic retrieval solutions mostly use language models to encode text into high-dimensional vectors and perform approximate k-nearest neighbor search for retrieval. However, the high cost of VectorDB and language model deployment has been a barrier for many users. AWS’s Neural Sparse Retrieval technology, on the other hand, achieves efficient semantic search while reducing costs.
It is understood that the technology was developed by Dr. Yang Yang, the machine learning leader of the OpenSearch China R&D team, as well as machine learning engineers Geng Zhichao and Guan Cong. With their deep academic background and practical experience, they have successfully developed this innovative technology.
Currently, this technology has received widespread attention and application worldwide. With over 500 million downloads and more than 70 enterprise partners, the technology’s strength and influence are proven.
AWS stated that it will continue to deepen its investment in AI, launching more innovative technologies to provide better services to users. At the same time, it welcomes more talented individuals to join their team and jointly promote the development of AI technology.
As an important platform for global academic communication and dissemination, the Machine Heart AIxiv column will continue to pay close attention to and report on the latest progress of AWS in the field of AI. If you have excellent work to share, please feel free to submit or contact us for coverage.
It is believed that with the efforts of AWS and other enterprises, semantic retrieval technology will continue to progress, bringing more convenience and intelligent experiences to people’s lives.
【来源】https://www.jiqizhixin.com/articles/2024-07-01-5
Views: 5