Search
Search documents
Layering every technique in RAG, one query at a time - David Karam, Pi Labs (fmr. Google Search)
AI Engineer· 2025-07-29 14:30
RAG技术栈 - RAG技术栈范围从最简单的内存嵌入和相关性排序搜索,到最复杂的行星级搜索,后者包含70多种语料库混合,包括token、embeddings和知识图谱[1] - 行业正在探索在200毫秒内以每秒16万次查询的速度,对这些混合语料库进行联合检索、自定义排序、联合重排序和LLM处理[1] - 报告通过“一次一个查询”的方式,逐步增加复杂性,展示RAG中所有技术的局限性,以及下一层技术在处理更复杂查询方面的能力[1] 搜索挑战 - 某些搜索问题非常难以解决,以至于行业可能更倾向于将问题交给LLM或UX处理[1] - 报告指出,像[falafel]这样的查询非常难以搜索,而对文档进行分块可能会是灾难性的[1] 行业应用与洞察 - Google团队在50多个搜索产品(包括Google.com和定制企业搜索)的背景下,分享了RAG技术的应用经验[1] - Pi Labs 致力于将Google在搜索核心AI和NLU系统方面的工作经验带给整个行业[1]
开源套壳叫板Google?Perplexity新品发布,印度裔CEO放言5万美金撬走彭博千亿生意
AI前线· 2025-07-22 09:32
Core Viewpoint - Perplexity has launched a new web browser named Comet, aiming to challenge Google Chrome's dominance in the market, which currently holds a 66.6% market share. The launch coincides with rumors of OpenAI's own browser release, indicating a competitive landscape in AI-driven search and browsing tools [1][2][3]. Group 1: Product Launch and Market Strategy - Comet integrates Perplexity's AI search tools and smart assistant to enhance user experience, initially available to premium users at $200 per month [1]. - Perplexity's ambition extends beyond user acquisition; they aim to replicate and potentially surpass Google's business model [1][2]. - The company has expressed willingness to acquire Google Chrome if legal pressures force Google to divest it, indicating a strategic move to capture a larger market share [1]. Group 2: Data Acquisition and Advertising - CEO Aravind Srinivas highlighted the importance of gathering user behavior data outside of Perplexity's applications to improve advertising quality, framing the browser as part of a broader data strategy [2]. - The decision to create a browser stemmed from a rejection by Google to include Perplexity as a default search engine, prompting the company to develop its own solution [2][3]. Group 3: Competitive Landscape and Industry Commentary - The Comet browser is built on Google's open-source Chromium project, which raises questions about the originality of Perplexity's offering [3]. - The current trend in startups involves forking open-source projects to add paid features, reflecting a broader entrepreneurial strategy in the tech industry [4]. Group 4: Vision and Long-term Goals - Perplexity aims to leverage AI to transform decision-making processes in finance, targeting a market valued at tens of trillions of dollars, significantly larger than Google's annual revenue [8][36]. - The company seeks to disrupt the financial research market dominated by Bloomberg, proposing that AI can enhance decision-making efficiency and democratize access to financial insights [36][37]. Group 5: Product Features and User Experience - Comet is envisioned as a "cognitive operating system" that integrates AI into daily workflows, allowing users to automate tasks and improve productivity [14][15]. - The browser will enable users to issue commands directly, with AI handling tasks such as data extraction and report generation, enhancing the overall user experience [15][34]. Group 6: Funding and Investor Relations - Perplexity's development costs were notably low, with the product being built for just $50,000, which impressed investors like Marc Andreessen [7][28]. - The company has faced skepticism from investors who prefer safer, vertical market strategies, but Srinivas remains committed to tackling larger, more challenging problems [6][26].
Recsys Keynote: Improving Recommendation Systems & Search in the Age of LLMs - Eugene Yan
AI Engineer· 2025-07-16 15:00
Industry Trend - Recommendation and search systems have been significantly impacted by advances in language modeling, evolving from Word2vec to GRUs, Transformers, and BERT [1] - The emergence of large language models (LLMs) is driving innovation in model architecture, scalable system designs, and customer experiences within recommendation and search systems [1] - The industry is exploring real-world implementations and measurable outcomes of LLMs in recommendation and search systems [1] Technological Advancement - LLM-driven techniques are expected to shape the future of content discovery and intelligent search [1] - Amazon is building recommendation systems and AI-powered products using ML/AI [1]
开源套壳叫板Google?Perplexity新品发布,印度裔CEO放言5万美金撬走彭博千亿生意
Sou Hu Cai Jing· 2025-07-12 03:31
Core Insights - Perplexity has launched a new web browser named Comet, integrating its AI search tools and smart assistant to provide a more intelligent browsing experience, initially available to premium users at $200 per month [1] - The launch of Comet is a direct challenge to Google Chrome, which holds a market share of 66.6%, and comes at a time when rumors suggest OpenAI is also developing its own AI browser [1][2] - Perplexity aims not only to compete with Google for users but also to replicate and potentially surpass Google's business model [1][2] Company Strategy - The U.S. Department of Justice has pressured Google to sell Chrome and end its default search engine agreements, prompting Perplexity to express interest in acquiring Chrome if such a ruling occurs [2] - CEO Aravind Srinivas has indicated that the development of their own browser is also aimed at gathering user behavior data outside of the Perplexity app to enhance advertising quality [2] - Srinivas has criticized the conventional wisdom in the startup community that suggests focusing on niche markets to avoid competition with giants like Google, asserting that he prefers to tackle significant challenges that attract top engineering talent [4][5] Market Ambitions - Perplexity's ambitions extend beyond search; the company aims to disrupt the financial information market, specifically targeting Bloomberg, by leveraging AI to enhance decision-making processes [6][27] - Srinivas estimates that the daily decision-making value in the financial system is in the tens of trillions of dollars, and even capturing 1% of that could yield $100 billion daily, far exceeding Google's annual revenue of approximately $200 billion [6][27] - The company is positioning itself to transform the financial research landscape by offering a more efficient and user-friendly alternative to traditional tools like Bloomberg Terminal [28] Product Features - Comet is envisioned as a "cognitive operating system" that integrates AI into the browsing experience, allowing users to automate tasks and streamline workflows [9][10] - The browser will enable users to issue commands directly, such as finding specific job candidates or managing tasks, with AI handling the execution [9][10] - Perplexity Labs, a feature of the platform, allows users to conduct in-depth research and generate content, further enhancing the utility of the browser [25][26] Competitive Landscape - The launch of Comet and the focus on AI integration in browsing highlight a growing trend in the tech industry where companies are seeking to redefine user interaction with web tools [1][9] - Perplexity's approach contrasts with traditional models that rely heavily on human analysts, suggesting a shift towards AI-driven solutions in financial decision-making [28] - The company is also addressing concerns about content creators by implementing a publisher program to ensure that quality information sources are compensated fairly, differentiating itself from competitors like Google [30]
X @The Wall Street Journal
The Wall Street Journal· 2025-07-05 11:48
A new crop of startups are betting on the rapid demise of traditional Google search https://t.co/VuAbs8zMaW ...
Over a Third of US Population Uses GenAI
Bloomberg Technology· 2025-06-30 19:46
Generative AI Adoption Trends - Generative AI tool usage is expected to be similar between Gen Z and Millennials by 2025 [1] - By 2029, the majority of internet users are projected to use generative AI tools [2] - Over 75% of Gen Z internet users are expected to use generative AI tools [3] Market Share and Competition - Chat GPT currently dominates the generative AI market, used by approximately 75% of generative AI users [4] - While Chat GPT's absolute user numbers are increasing, its share of generative AI users is decreasing due to growing competition [5] - Gemini and Microsoft Copilot are increasing competition in the generative AI market [5] Generative AI Integration - Many individuals are using generative AI through integrated platforms like Google Search without actively seeking dedicated apps [6][7] - Generative AI is becoming pervasive in everyday activities, including shopping, chatbots, AI overviews, and customer service [8] Use Cases - Search is the leading use case for generative AI, currently used by approximately 64% of users and expected to increase to 75% [9] - Generative AI is also used for work-related tasks like coding and productivity tools by approximately 64% to 74% of users [10] - Personal recommendations, such as shopping tools, are used by around 40% of people [11]
A quick guide | How to search in ChatGPT
OpenAI· 2025-06-27 19:51
Search Experience Enhancement - Introduces a new search experience within Chad GPT, contrasting with traditional keyword-based internet searches [1] - Emphasizes the provision of overviews with citations, facilitating deeper exploration of information [2][4] Use Cases and Model Selection - Highlights the utility of Chad GPT for researching topics involving large amounts of information, exemplified by student study habits on Reddit [2] - Recommends the O3 model for light research purposes [3] - Suggests using the default model for queries that may or may not require up-to-date information, allowing the model to decide [4] - Offers the option to explicitly select up-to-date information by clicking the globe icon [5] Practical Applications - Showcases practical applications such as finding events in San Francisco museums and co-working spaces with day passes [4] User Feedback - Encourages users to try searching in Chad GPT and provide feedback [5]
Security Is The New Search: Alphabet's $32 Billion Wiz Deal Explained
Seeking Alpha· 2025-03-19 14:25
Alphabet Inc. (NASDAQ: GOOG ) , which is Google’s parent company, is probably the only company I can recall both the year I first heard about it and the person who told me. Back then, I had tried Ask Jeeves, remember that? Nice logo, catchy name, didn’t really work, but, likeThe mission of Grassroots Trading rests on the following principles: providing objective, unbiased, and balanced research, backed by solid data and completely void of emotional influences or preference for companies; focusing on small- ...