AI knowledge

Large-scale language modeling (LLM) is playing an increasingly important role in the field of artificial intelligence. In order to better understand and apply LLMs, we need to gain a deeper understanding of their core concepts. In this paper, we will focus on three key concepts, namely Token, Maximum Output Length, and Context Length, to help readers clear the understanding barriers so as to...

Recently, the terms Autonomous AI (AI), AI Agents, and Agents have been popping up a lot. Frankly, despite being data analysts and scientists, industry players have been a bit resistant to these AI-related trends and buzzwords in the past...

In recent years, Artificial Intelligence (AI) technologies have triggered a profound change in the field of programming. From v0 and bolt.new to programming tools that integrate Agent technology such as Cursor and Windsurf, AI Coding shows great potential to play a key role in the software development process, especially in rapid proto...

In the age of AI-assisted programming, we want AI to generate code that is not just static text, but can be parsed, edited, previewed, and even executed. This demand has given rise to a new interaction paradigm - Artifact. In this article, we will analyze Artifact from theoretical concepts to practical implementation....

In this paper, we present a summary report of Kapa.ai's recent exploration of OpenAI's o3-mini and other inference models in the Retrieval-Augmented Generation (RAG) system. Kapa.ai is an AI assistant powered by a large-scale language model (LLM) that...

Recently in the intelligent customer service project to choose the RAG knowledge base of data processing tools, it re-looked at the current mainstream document processing projects, including olmOCR, Marker, MinerU, Docling, Markitdown, Llamaparse the 6 tools, and a brief comparison of them. A comprehensive view of the ...

DeepSeek R1 has demonstrated strong inference capabilities in its first release. In this blog post, we share in detail our experience using DeepSeek R1 to build a Retrieval-Augmented Generation (RAG) system, specifically for the legal document domain. We chose ...

Abstract Well-designed prompts are essential to enhance the reasoning capabilities of large language models (LLMs) while aligning their outputs with the task requirements of different domains. However, manually designing hints requires expertise and iterative experimentation. Existing hint optimization methods aim to automate this process, but they heavily ...

It's only February, and Deep Search is already looming as the new search standard for 2025. Giants like Google and OpenAI have unveiled their "Deep Research" products in an effort to capitalize on this wave of technology. (We're proud to be releasing our...

By Krish Maniar and William Fu-Hinthorn When writing cue words, we try to communicate our intentions to Large Language Models (LLMs) so that they can apply these instructions on complex data. However, it is not easy to clearly express all the nuances at once. Prompts are usually engineered by hand ...

This diagram clearly depicts the architectural blueprint of a modern, complex Question Answering (QA) or Retrieval-Augmented Generation (RAG) system. It begins with the user asking the question and continues through the final generation of the answer, showing in detail the intermediate...

At the center of the wave of artificial intelligence, AI Agent (Intelligent Body) is evolving at an amazing speed, just like the intelligent assistants coming out of sci-fi movies, quietly penetrating into every corner of the enterprise. They are no longer unattainable future concepts, but the secret weapon for enterprises to improve efficiency, optimize processes, and win in the market. 2...

Amidst the ever-changing wave of translation technologies, the emergence of ChatGPT (Chat Generative Pre-trained Transformer) has undoubtedly attracted global attention. As a state-of-the-art Large Language Models (LLM), ChatGPT demonstrates impressive natural language...

After OpenAI's Deep Research tool came out of nowhere, all the major vendors launched their own Deep Research tools. The so-called Deep Research is compared with ordinary search, where a simple RAG search generates generally only one round of search. However Deep Research can act like a human, based on a...

Technology Core: Retrieval Interleaved Generation (RIG) What is RIG? RIG is an innovative generation methodology designed to address the problem of hallucination in the processing of statistical data by large language models. Traditional models may generate inaccurate numbers or facts out of thin air, while...

If your RAG application is failing to deliver the desired results, perhaps it's time to revisit your chunking strategy. Better chunking means more accurate searches and, ultimately, higher quality responses. However, chunking is not a one-size-fits-all technique, and no single approach is absolutely optimal. You'll need to tailor your...