[Docs] Update blog: In-Context-Search#9
Conversation
Summary of ChangesHello @wangxingjun778, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request refines the 'In-Context Search' blog post by updating its publication date and enhancing its content with direct links to the discussed projects. It also introduces a comprehensive references section, providing readers with further resources and academic context. A minor structural correction was made to the Chinese translation. Highlights
🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console. Changelog
Activity
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
The pull request updates blog posts related to 'In-Context Search' in both English and Chinese versions. The changes include updating the date, refining introductory text, adding GitHub links to project mentions (PageIndex, Sirchmunk, LightRAG), and incorporating a new 'References' section at the end of both articles. The changes improve the clarity and provide valuable external links for readers.
| --- | ||
|
|
||
| Traditional Retrieval-Augmented Generation (RAG) frameworks have established a robust foundation for grounding Large Language Models (LLMs) in external knowledge through static indexing and vector similarity. However, as computational paradigms shift toward **LLM-native** architectures, a new frontier known as **In-Context Search (ICS)** is emerging. This post analyzes the transition from traditional Graph-based RAG to next-generation ICS paradigms, represented by **VectifyAI's PageIndex** and **ModelScope's Sirchmunk**. | ||
| With the evolution of RAG (Retrieval-Augmented Generation) technology, a new paradigm called **In-Context Search (ICS)** is redefining how LLMs interact with external knowledge. This post compares traditional Graph-based RAG with next-generation ICS approaches represented by **[PageIndex](https://github.com/VectifyAI/PageIndex)** and **[Sirchmunk](https://github.com/modelscope/sirchmunk)**. |
There was a problem hiding this comment.
The introduction has been rephrased for better flow and clarity. Adding direct links to the GitHub repositories for PageIndex and Sirchmunk is a good enhancement, providing immediate access to the mentioned projects.
| ## 1. The Foundation and Frontiers of RAG | ||
|
|
||
| The first generation of RAG successfully addressed LLM hallucinations by introducing external knowledge bases. These systems typically rely on **Vector Databases** or **Static Knowledge Graphs** (e.g., **LightRAG**). | ||
| The first generation of RAG successfully addressed LLM hallucinations by introducing external knowledge bases. These systems typically rely on **Vector Databases** or **Static Knowledge Graphs** (e.g., **[LightRAG](https://github.com/HKUDS/LightRAG)**). |
There was a problem hiding this comment.
Adding a GitHub link for LightRAG here is consistent with the previous change and provides useful context for readers interested in the project.
|
|
||
| --- | ||
|
|
||
| ## References | ||
|
|
||
| 1. Lewis, P., Perez, E., Piktus, A., et al. (2020). *Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks.* NeurIPS 2020. [arXiv:2005.11401](https://arxiv.org/abs/2005.11401) | ||
| 2. Guo, Z., Qian, C., et al. (2024). *LightRAG: Simple and Fast Retrieval-Augmented Generation.* [arXiv:2410.05779](https://arxiv.org/abs/2410.05779) | [GitHub](https://github.com/HKUDS/LightRAG) | ||
| 3. VectifyAI. (2025). *PageIndex: Extracting and Understanding Financial Reports with LLM.* [GitHub](https://github.com/VectifyAI/PageIndex) | ||
| 4. ModelScope. (2025). *Sirchmunk: An Embedding-Free, Agentic Search Engine for Raw Data.* [GitHub](https://github.com/modelscope/sirchmunk) | ||
| 5. Yao, S., Zhao, J., Yu, D., et al. (2023). *ReAct: Synergizing Reasoning and Acting in Language Models.* ICLR 2023. [arXiv:2210.03629](https://arxiv.org/abs/2210.03629) | ||
| 6. Anthropic. (2024). *Model Context Protocol (MCP) Specification.* [Documentation](https://modelcontextprotocol.io) | ||
| 7. Kaddour, J., Harris, J., Mozes, M., et al. (2023). *Challenges and Applications of Large Language Models.* [arXiv:2307.10169](https://arxiv.org/abs/2307.10169) |
There was a problem hiding this comment.
The addition of a comprehensive 'References' section significantly improves the academic rigor and credibility of the blog post. It allows readers to delve deeper into the cited works.
| --- | ||
|
|
||
| 传统的检索增强生成（RAG）框架通过静态索引与向量相似度匹配，为大语言模型（LLM）构建了坚实的外部知识接入基础。然而，随着计算范式向 **LLM 原生** 架构演进，一种被称为 **上下文搜索（In-Context Search, ICS）** 的新前沿正在兴起。本文分析了从传统 Graph-based RAG 到下一代 ICS 范式的技术演进，以 **VectifyAI 的 PageIndex** 和 **ModelScope 的 Sirchmunk** 为代表。 | ||
| 随着 RAG (Retrieval-Augmented Generation) 技术的演进，一种名为 **上下文搜索（In-Context Search, ICS）** 的新范式正在重新定义 LLM 与外部知识的交互方式。本文对比了传统 Graph-based RAG 与以 **PageIndex** 和 **Sirchmunk** 为代表的下一代 ICS 方案。 |
There was a problem hiding this comment.
The Chinese version also benefits from the rephrased introduction, making it more concise and direct. Consider adding the GitHub links for PageIndex and Sirchmunk here as well, similar to the English version, for consistency and reader convenience.
References
- Ensure consistency across different language versions of the same content, especially for external links or key references.
|
|
||
|
|
||
| ## 5. 挑战与未来研究方向 | ||
| ## 4. 挑战与未来研究方向 |
There was a problem hiding this comment.
The section heading has been updated from '5. 挑战与未来研究方向' to '4. 挑战与未来研究方向'. Please ensure that the numbering of all sections throughout the document is consistent and logical after this change.
| --- | ||
|
|
||
| ## 参考文献 | ||
|
|
||
| 1. Lewis, P., Perez, E., Piktus, A., 等. (2020). *Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks.* NeurIPS 2020. [arXiv:2005.11401](https://arxiv.org/abs/2005.11401) | ||
| 2. Guo, Z., Qian, C., 等. (2024). *LightRAG: Simple and Fast Retrieval-Augmented Generation.* [arXiv:2410.05779](https://arxiv.org/abs/2410.05779) | [GitHub](https://github.com/HKUDS/LightRAG) | ||
| 3. VectifyAI. (2025). *PageIndex: Extracting and Understanding Financial Reports with LLM.* [GitHub](https://github.com/VectifyAI/PageIndex) | ||
| 4. ModelScope. (2025). *Sirchmunk：一个无嵌入的、智能体驱动的原始数据搜索引擎。* [GitHub](https://github.com/modelscope/sirchmunk) | ||
| 5. Yao, S., Zhao, J., Yu, D., 等. (2023). *ReAct: Synergizing Reasoning and Acting in Language Models.* ICLR 2023. [arXiv:2210.03629](https://arxiv.org/abs/2210.03629) | ||
| 6. Anthropic. (2024). *模型上下文协议（MCP）规范。* [官方文档](https://modelcontextprotocol.io) | ||
| 7. Kaddour, J., Harris, J., Mozes, M., 等. (2023). *Challenges and Applications of Large Language Models.* [arXiv:2307.10169](https://arxiv.org/abs/2307.10169) |
There was a problem hiding this comment.
The addition of the '参考文献' (References) section in the Chinese version is a valuable improvement, providing academic backing and further reading for the audience.
1 participant
No description provided.