# \#embeddings **URL:** https://www.artofsm.art/tag/embeddings/18.md [Latest](https://www.artofsm.art/latest.md) · [Categories](https://www.artofsm.art/categories.md) · [Tags](https://www.artofsm.art/tags.md) --- ## [INSANE: This 46 MB AI Model Runs 5x FASTER on My Laptop (Frozen Embeddings TESTED)](https://www.artofsm.art/t/insane-this-46-mb-ai-model-runs-5x-faster-on-my-laptop-frozen-embeddings-tested/26507)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [20 October 2026 18:24 UTC](https://www.artofsm.art/t/insane-this-46-mb-ai-model-runs-5x-faster-on-my-laptop-frozen-embeddings-tested/26507 "2026-10-20T18:24:05Z")
The video demonstrates that using a frozen, untrained embedding table significantly reduces the size and increases the speed of small language models like Gemma 3, while maintaining or even improving perfor… --- ## [This 150M Model Claims #1. I Tested It](https://www.artofsm.art/t/this-150m-model-claims-1-i-tested-it/26022)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [16 October 2026 00:54 UTC](https://www.artofsm.art/t/this-150m-model-claims-1-i-tested-it/26022 "2026-10-16T00:54:41Z")
The video reviews the Ultraert Embed 150M, a compact embedding model designed for long-context tasks and retrieval, highlighting its strong performance on long documents and code search but moderate results… --- ## [Judge a Transaction by the Company It Keeps — Somya Rai & MD Sharique, EXL](https://www.artofsm.art/t/judge-a-transaction-by-the-company-it-keeps-somya-rai-md-sharique-exl/25815)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [13 October 2026 13:24 UTC](https://www.artofsm.art/t/judge-a-transaction-by-the-company-it-keeps-somya-rai-md-sharique-exl/25815 "2026-10-13T13:24:16Z")
Somya Rai and MD Sharique from EXL present a fraud detection approach that combines graph neural networks with traditional machine learning to analyze the relational context of transactions, significantly i… --- ## [Vector Isn't Enough: Hybrid Search & Retrieval — Jeff Vestal & James Williams, Elastic](https://www.artofsm.art/t/vector-isnt-enough-hybrid-search-retrieval-jeff-vestal-james-williams-elastic/25659)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [11 October 2026 13:24 UTC](https://www.artofsm.art/t/vector-isnt-enough-hybrid-search-retrieval-jeff-vestal-james-williams-elastic/25659 "2026-10-11T13:24:40Z")
The workshop by Jeff Vestal and James Williams from Elastic explores hybrid search by combining semantic vector search with traditional lexical methods to enhance retrieval accuracy and relevance, supported… --- ## [Google's Tiny Embedding Model Now Does Audio, Video and Images](https://www.artofsm.art/t/googles-tiny-embedding-model-now-does-audio-video-and-images/25198)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [7 October 2026 12:54 UTC](https://www.artofsm.art/t/googles-tiny-embedding-model-now-does-audio-video-and-images/25198 "2026-10-07T12:54:39Z")
Google’s Embedding Gemini 2 is a versatile, open-source multimodal embedding model that integrates text, images, audio, and video into a unified vector space, enabling accurate cross-modal search and retrie… --- ## [Grep or Embeddings? Agentic Search Over Company Documents — George He, LlamaIndex](https://www.artofsm.art/t/grep-or-embeddings-agentic-search-over-company-documents-george-he-llamaindex/25172)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [7 October 2026 01:55 UTC](https://www.artofsm.art/t/grep-or-embeddings-agentic-search-over-company-documents-george-he-llamaindex/25172 "2026-10-07T01:55:10Z")
In his talk, George He from LlamaIndex presents a hybrid search approach that combines traditional keyword-based and embedding-based semantic vector search to efficiently manage and retrieve information fro… --- ## [AI & Data Science Periodic Tables: How They Work Together](https://www.artofsm.art/t/ai-data-science-periodic-tables-how-they-work-together/22445)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [13 August 2026 11:20 UTC](https://www.artofsm.art/t/ai-data-science-periodic-tables-how-they-work-together/22445 "2026-08-13T11:20:37Z")
The video highlights the interdependent relationship between data science and AI, demonstrating how data preparation and governance underpin AI models while AI tools increasingly automate and enhance data s… --- ## [What Is AI Model Collapse? Why AI Could Forget Reality](https://www.artofsm.art/t/what-is-ai-model-collapse-why-ai-could-forget-reality/22175)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [6 August 2026 11:20 UTC](https://www.artofsm.art/t/what-is-ai-model-collapse-why-ai-could-forget-reality/22175 "2026-08-06T11:20:57Z")
Model collapse is a phenomenon where AI models trained repeatedly on AI-generated data gradually lose rare and accurate real-world information, leading to increasingly generic, less reliable outputs that dr… --- ## [What Are Large Database Models? AI for SQL Data](https://www.artofsm.art/t/what-are-large-database-models-ai-for-sql-data/22094)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [4 August 2026 11:24 UTC](https://www.artofsm.art/t/what-are-large-database-models-ai-for-sql-data/22094 "2026-08-04T11:24:42Z")
Large Database Models (LDMs) are AI models specifically designed to analyze relational databases by learning patterns directly from data tables, enabling advanced semantic queries like similarity searches a… --- ## [Stanford CS229 Machine Learning | Spring 2026 | Lecture 13: LLMs, Next-Word Prediction Loss](https://www.artofsm.art/t/stanford-cs229-machine-learning-spring-2026-lecture-13-llms-next-word-prediction-loss/21974)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [31 July 2026 22:39 UTC](https://www.artofsm.art/t/stanford-cs229-machine-learning-spring-2026-lecture-13-llms-next-word-prediction-loss/21974 "2026-07-31T22:39:31Z")
The lecture covers representation learning through embeddings, highlighting supervised and contrastive learning methods to create vector representations that enable semantic search and similarity-based task… --- ## [Quantum Machine Learning - Computerphile](https://www.artofsm.art/t/quantum-machine-learning-computerphile/21579)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [23 July 2026 13:25 UTC](https://www.artofsm.art/t/quantum-machine-learning-computerphile/21579 "2026-07-23T13:25:04Z")
The video explains how quantum machine learning leverages quantum computing principles like superposition and entanglement to embed classical data into higher-dimensional spaces, enabling potentially more e… --- ## [AI on Your Lakehouse: Context Comes in Shapes, Not Queries — Zach Blumenfeld, Neo4j](https://www.artofsm.art/t/ai-on-your-lakehouse-context-comes-in-shapes-not-queries-zach-blumenfeld-neo4j/21564)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [23 July 2026 07:24 UTC](https://www.artofsm.art/t/ai-on-your-lakehouse-context-comes-in-shapes-not-queries-zach-blumenfeld-neo4j/21564 "2026-07-23T07:24:11Z")
In this workshop, Zach Blumenfeld from Neo4j introduces participants to the concept of “AI on Your Lakehouse,” emphasizing the use of contextual data shapes over traditional queries to enhance AI applicatio… --- ## [Video Has No Memory. Here's How We Built One. — James Le, TwelveLabs](https://www.artofsm.art/t/video-has-no-memory-heres-how-we-built-one-james-le-twelvelabs/21556)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [23 July 2026 02:23 UTC](https://www.artofsm.art/t/video-has-no-memory-heres-how-we-built-one-james-le-twelvelabs/21556 "2026-07-23T02:23:36Z")
James Le from TwelveLabs highlights the limitations of current video AI systems that treat videos as isolated frames or text tokens, lacking true memory and reasoning across time and modalities. To overcome… --- ## [A Practitioner's Guide to Graphs - Tim Ainge, Good Collective](https://www.artofsm.art/t/a-practitioners-guide-to-graphs-tim-ainge-good-collective/21375)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [18 July 2026 21:50 UTC](https://www.artofsm.art/t/a-practitioners-guide-to-graphs-tim-ainge-good-collective/21375 "2026-07-18T21:50:26Z")
Tim Ainge from The Good Collective presents a practitioner’s guide to graphs, explaining how well-structured graph data—built using schemas, standardized data, and entity matching—can enhance AI application… --- ## [Extreme Token Use of Agentic AI - Computerphile](https://www.artofsm.art/t/extreme-token-use-of-agentic-ai-computerphile/20785)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [2 July 2026 14:47 UTC](https://www.artofsm.art/t/extreme-token-use-of-agentic-ai-computerphile/20785 "2026-07-02T14:47:13Z")
The video explains that agentic AI coding tools incur high costs due to the massive token usage required to process large contexts and repeated interactions, as each new token generation involves reprocessi… --- ## [We Cut 94% of AI Coding Tokens With a Local Code Index - Rajkumar Sakthivel, Tesco](https://www.artofsm.art/t/we-cut-94-of-ai-coding-tokens-with-a-local-code-index-rajkumar-sakthivel-tesco/20656)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [28 June 2026 22:50 UTC](https://www.artofsm.art/t/we-cut-94-of-ai-coding-tokens-with-a-local-code-index-rajkumar-sakthivel-tesco/20656 "2026-06-28T22:50:49Z")
Rajkumar Sakthivel and his colleague Foss developed a local code indexing system that drastically reduces AI coding costs by minimizing the input tokens sent to AI models through semantic and keyword search… --- ## [When All Context Matters: Extended Cache Augmented Generation - Luis Romero-Sevilla, Orbis](https://www.artofsm.art/t/when-all-context-matters-extended-cache-augmented-generation-luis-romero-sevilla-orbis/20648)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [28 June 2026 20:20 UTC](https://www.artofsm.art/t/when-all-context-matters-extended-cache-augmented-generation-luis-romero-sevilla-orbis/20648 "2026-06-28T20:20:33Z")
Luis Romero-Sevilla presents extended Cache Augmented Generation (CAG) as a scalable solution for handling rapidly changing, large document collections where all context is relevant, by distributing documen… --- ## [Turbocharge Your Agent's Retrieval with TurboQuant - Shashi Jagtap, Superagentic AI](https://www.artofsm.art/t/turbocharge-your-agents-retrieval-with-turboquant-shashi-jagtap-superagentic-ai/20643)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [28 June 2026 18:20 UTC](https://www.artofsm.art/t/turbocharge-your-agents-retrieval-with-turboquant-shashi-jagtap-superagentic-ai/20643 "2026-06-28T18:20:37Z")
Shashi Jagtap from Super Agentic AI presents TurboQuant, a novel compression algorithm that reduces memory usage in agent retrieval systems by compressing embeddings to 3-4 bits without sacrificing search q… --- ## [The 100-Tool Agent Is a Trap - Sohail Shaikh & Ankush Rastogi, Prosodica](https://www.artofsm.art/t/the-100-tool-agent-is-a-trap-sohail-shaikh-ankush-rastogi-prosodica/20636)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [28 June 2026 15:35 UTC](https://www.artofsm.art/t/the-100-tool-agent-is-a-trap-sohail-shaikh-ankush-rastogi-prosodica/20636 "2026-06-28T15:35:41Z")
Sohail Shaikh and Ankush Rastogi highlight the pitfalls of the “100-tool agent” approach, where loading all tool schemas simultaneously leads to poor accuracy, high latency, and increased costs, especially … --- ## [World Models - The Next Big AI Revolution | AI Animated](https://www.artofsm.art/t/world-models-the-next-big-ai-revolution-ai-animated/20236)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [15 June 2026 15:39 UTC](https://www.artofsm.art/t/world-models-the-next-big-ai-revolution-ai-animated/20236 "2026-06-15T15:39:10Z")
The video explains how traditional AI models like large language models lack true understanding and internal simulation of the world, leading to limitations in reasoning and planning. It introduces world mo… --- ## [RAG is dead, right? — Kuba Rogut, Turbopuffer](https://www.artofsm.art/t/rag-is-dead-right-kuba-rogut-turbopuffer/20014)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [9 June 2026 17:20 UTC](https://www.artofsm.art/t/rag-is-dead-right-kuba-rogut-turbopuffer/20014 "2026-06-09T17:20:54Z")
Kuba Rogut from Turbo Puffer argues that retrieval-augmented generation (RAG) is not dead but evolving into more sophisticated agentic search methods that combine various retrieval techniques and iterative … --- ## [Benchmarking semantic code retrieval on Claude Code — Kuba Rogut, Turbopuffer](https://www.artofsm.art/t/benchmarking-semantic-code-retrieval-on-claude-code-kuba-rogut-turbopuffer/19500)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [3 June 2026 15:20 UTC](https://www.artofsm.art/t/benchmarking-semantic-code-retrieval-on-claude-code-kuba-rogut-turbopuffer/19500 "2026-06-03T15:20:28Z")
Kuba Rogut from Turbo Puffer discusses benchmarking semantic code retrieval using Claude Code, highlighting how semantic search with vector embeddings significantly improves code retrieval accuracy and effi… --- ## [Yann LeCun's $1B Bet Against LLMs \[Part 2\]](https://www.artofsm.art/t/yann-lecuns-1b-bet-against-llms-part-2/19373)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [30 May 2026 18:39 UTC](https://www.artofsm.art/t/yann-lecuns-1b-bet-against-llms-part-2/19373 "2026-05-30T18:39:24Z")
The video discusses Yann LeCun’s promotion of the JEPA (Joint Embedding Predictive Architecture) approach as a more scalable and robust alternative to mainstream Vision-Language-Action (VLA) models in AI, e… --- ## [The Complete Guide to Hybrid Search in RAG (BM25 + Embeddings + Reranker)](https://www.artofsm.art/t/the-complete-guide-to-hybrid-search-in-rag-bm25-embeddings-reranker/18767)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [14 May 2026 18:17 UTC](https://www.artofsm.art/t/the-complete-guide-to-hybrid-search-in-rag-bm25-embeddings-reranker/18767 "2026-05-14T18:17:20Z")
The tutorial guides viewers through building a hybrid retrieval system combining BM25, dense embeddings, reciprocal rank fusion, and a re-ranker to optimize document retrieval on the Financial QA dataset, h… --- ## [Give Your Chat Agent a Voice — Luke Harries, ElevenLabs](https://www.artofsm.art/t/give-your-chat-agent-a-voice-luke-harries-elevenlabs/18513)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [9 May 2026 13:20 UTC](https://www.artofsm.art/t/give-your-chat-agent-a-voice-luke-harries-elevenlabs/18513 "2026-05-09T13:20:28Z")
Luke Harries from ElevenLabs presents Voice Engine, a solution that easily adds natural, context-aware voice capabilities to existing chat agents without requiring extensive redevelopment, enhancing accessi… --- ## [RAG's Evolution: From Simple Retrieval to Agentic AI](https://www.artofsm.art/t/rags-evolution-from-simple-retrieval-to-agentic-ai/18355)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [5 May 2026 11:20 UTC](https://www.artofsm.art/t/rags-evolution-from-simple-retrieval-to-agentic-ai/18355 "2026-05-05T11:20:49Z")
The evolution of information retrieval has progressed from basic keyword-based search to sophisticated agentic Retrieval-Augmented Generation (RAG) AI, which combines large language models with dynamic retr… --- ## [Yann LeCun's Billion Dollar Bet](https://www.artofsm.art/t/yann-lecuns-billion-dollar-bet/18260)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [2 May 2026 02:05 UTC](https://www.artofsm.art/t/yann-lecuns-billion-dollar-bet/18260 "2026-05-02T02:05:25Z")
Yann LeCun has invested a billion dollars in developing JEPA, an AI approach that focuses on learning internal representations through joint embedding predictive architectures to build world models capable … --- ## [Learn How to Make AI Models w/ ML: 4. Embeddings & RAG](https://www.artofsm.art/t/learn-how-to-make-ai-models-w-ml-4-embeddings-rag/17924)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [23 April 2026 14:20 UTC](https://www.artofsm.art/t/learn-how-to-make-ai-models-w-ml-4-embeddings-rag/17924 "2026-04-23T14:20:28Z")
This episode explains how embeddings represent text meaning as vectors to enable semantic search and Retrieval-Augmented Generation (RAG), demonstrating their use with vector databases like FAISS for effici… --- ## [3 Ways to Do RAG Without Vector Databases (Explained Simply)](https://www.artofsm.art/t/3-ways-to-do-rag-without-vector-databases-explained-simply/17829)
**Author:** [@merefield](https://www.artofsm.art/u/merefield)\ **Replies:** 1\ **Last updated:** [20 April 2026 17:24 UTC](https://www.artofsm.art/t/3-ways-to-do-rag-without-vector-databases-explained-simply/17829 "2026-04-20T17:24:53Z")
The video explains that Retrieval Augmented Generation (RAG) can be effectively implemented without relying solely on vector databases by using alternatives like dynamic documents, file-based storage, or co… --- ## [This AI Might Be a PERFECT Research Companion (Recall 2026 Review)](https://www.artofsm.art/t/this-ai-might-be-a-perfect-research-companion-recall-2026-review/17594)
**Author:** [@artesia](https://www.artofsm.art/u/artesia)\ **Replies:** 1\ **Last updated:** [14 April 2026 08:21 UTC](https://www.artofsm.art/t/this-ai-might-be-a-perfect-research-companion-recall-2026-review/17594 "2026-04-14T08:21:19Z")
Recall is an AI-powered knowledge base that helps researchers efficiently manage and interact with diverse academic content through AI-generated summaries, chat-based queries, visual graph mapping, and spac… [Next page](https://www.artofsm.art/tag/embeddings/18.md?match_all_tags=true&page=1&tags%5B%5D=embeddings)