Build-your-own graph query algorithm with LlamaIndex + Mistral AI 🔎🕸️ You’re not limited to text-to-cypher or pure vector search when you build GraphRAG - you can combine both techniques or define your own custom graph query algorithm, as long as you have access to the underlying property graph store. Ravi Theja Desetty has a full video tutorial and notebook showing you how to do this - combine text-to-cypher + vector retrieval with Cohere reranking, and synthesize the final results with mistral-large. Video: https://lnkd.in/g43-W-FJ Mistral AI cookbook: https://lnkd.in/gW6yEqdj
LlamaIndex’s Post
More Relevant Posts
-
Excited about the implementation of Adaptive RAG by Soyeong Jeong et al., which skillfully selects the most effective RAG strategy based on the complexity of the query. This advancement represents a significant leap in the realm of adaptive AI, improving both efficiency and precision in information retrieval. It's a transformative development for AI consulting, ensuring that clients receive swift, accurate responses for simple inquiries and a detailed, methodical analysis for multifaceted ones. #AIConsulting #AdaptiveAI #InformationRetrieval #MachineLearning #AI #ArtificialIntelligence #TechnologyInnovation #RAG #QueryComplexity #LlamaIndex
Adaptive RAG 🧠🔎 Not all questions are created equal. RAG techniques that do well on simple questions will fail on complex multi-step ones; fancy agentic techniques that address the latter will be way slower for simple questions. Adaptive RAG is a cool paper by Soyeong Jeong et al. that dynamically selects the best RAG strategy based on the query complexity. To do this, the authors propose a classifier that can classify different complexity levels. Thanks to Ravi Theja Desetty, we’re excited to implement the core idea as a Mistral AI cookbook, thanks for our Router abstraction. Thanks Sophia Yang, Ph.D. for the reviews! Check out our LlamaPack: https://lnkd.in/g3XQmfDb Paper: https://lnkd.in/d6DbStSK
-
-
🎛 Fine tune Mistral AI with your own datasets using Argilla Check our new tutorial with a practical example using the OIG dataset. You'll learn to: 1. Prepare a FeedbackDataset in Argilla. 2. (Optional) Annotate instruction samples. 3. Set up the ArgillaTrainer for Supervised Finetuning, leveraging our TRL integration. 4. Perform inference using the fine tuned LLM. 5. Publish the resulting model and dataset on the Hugging Face Hub. Link in the first comment
-
-
Adaptive RAG 🧠🔎 Not all questions are created equal. RAG techniques that do well on simple questions will fail on complex multi-step ones; fancy agentic techniques that address the latter will be way slower for simple questions. Adaptive RAG is a cool paper by Soyeong Jeong et al. that dynamically selects the best RAG strategy based on the query complexity. To do this, the authors propose a classifier that can classify different complexity levels. Thanks to Ravi Theja Desetty, we’re excited to implement the core idea as a Mistral AI cookbook, thanks for our Router abstraction. Thanks Sophia Yang, Ph.D. for the reviews! Check out our LlamaPack: https://lnkd.in/g3XQmfDb Paper: https://lnkd.in/d6DbStSK
-
-
Staff Engineer @ Office of CTO Jeavio | R&D Team Lead | Applied AI | Mentoring | Architecture and system Design
✅ Adaptive RAG ✅ As the saying goes "There is no silver bullet for all the problems". The same applies in the case of RAG as well. To find out the best strategy for the solution that we implement, we might experiment with different strategies and choose from that. There is also an approach for agent implementation which makes the whole process slow. There was a paper published recently about Adaptive RAG, which uses a classifier to decide the complexity level of query and based on that decides the strategy. i am excited to explore the possibility of optimal handling and dynamic selection to improve the performance and accuracy of responses using RAG. #RAG #llms #adaptiveRAG #GenAI #llamaindex
Adaptive RAG 🧠🔎 Not all questions are created equal. RAG techniques that do well on simple questions will fail on complex multi-step ones; fancy agentic techniques that address the latter will be way slower for simple questions. Adaptive RAG is a cool paper by Soyeong Jeong et al. that dynamically selects the best RAG strategy based on the query complexity. To do this, the authors propose a classifier that can classify different complexity levels. Thanks to Ravi Theja Desetty, we’re excited to implement the core idea as a Mistral AI cookbook, thanks for our Router abstraction. Thanks Sophia Yang, Ph.D. for the reviews! Check out our LlamaPack: https://lnkd.in/g3XQmfDb Paper: https://lnkd.in/d6DbStSK
-
-
User queries in general can be complex queries, simple queries. One don't always need complex RAG system even to handle simple queries. Adaptive RAG proposes an approach to handle complex queries and simple queries. Notebook : https://lnkd.in/gaPipwhe
Adaptive RAG 🧠🔎 Not all questions are created equal. RAG techniques that do well on simple questions will fail on complex multi-step ones; fancy agentic techniques that address the latter will be way slower for simple questions. Adaptive RAG is a cool paper by Soyeong Jeong et al. that dynamically selects the best RAG strategy based on the query complexity. To do this, the authors propose a classifier that can classify different complexity levels. Thanks to Ravi Theja Desetty, we’re excited to implement the core idea as a Mistral AI cookbook, thanks for our Router abstraction. Thanks Sophia Yang, Ph.D. for the reviews! Check out our LlamaPack: https://lnkd.in/g3XQmfDb Paper: https://lnkd.in/d6DbStSK
-
-
AI Engineer & Architect, Product Designer, Stable Diffusion Artist, Photographer & Melophile.
In a recent study, researchers have crafted a unique system that optimizes the way large language models respond to inquiries. The system stands out because of its adaptability, capable of addressing a wide range of questions from simple to complex with remarkable efficiency. (<-- EFFICIENCY) It utilizes a smaller model to gauge the complexity of an incoming question, then employs the most suitable method to answer it. According to comprehensive tests on various data sets, this approach significantly improves both the accuracy and efficiency of responses, outperforming existing methodologies. Right Tools, Right Job I always say. . #opensource #llamaindex #llms #mistral #ai #genai #generativeai #llamahub .
Adaptive RAG 🧠🔎 Not all questions are created equal. RAG techniques that do well on simple questions will fail on complex multi-step ones; fancy agentic techniques that address the latter will be way slower for simple questions. Adaptive RAG is a cool paper by Soyeong Jeong et al. that dynamically selects the best RAG strategy based on the query complexity. To do this, the authors propose a classifier that can classify different complexity levels. Thanks to Ravi Theja Desetty, we’re excited to implement the core idea as a Mistral AI cookbook, thanks for our Router abstraction. Thanks Sophia Yang, Ph.D. for the reviews! Check out our LlamaPack: https://lnkd.in/g3XQmfDb Paper: https://lnkd.in/d6DbStSK
-
-
Service Operations and Project Leader | Growth, Manufacturing, Digital Transformation | Cost Reduction
Not All Chatbots Are Created Equal: Adaptive RAG to the Rescue! Imagine a virtual assistant that truly understands your questions, regardless of complexity. That's the promise of Adaptive RAG, a groundbreaking approach by Soyeong Jeong. Unlike traditional chatbots, Adaptive RAG takes context and complexity into account, selecting the most effective strategy for crafting a precise and informative response. Intrigued? #AI #Chatbots #AdaptiveLearning
Adaptive RAG 🧠🔎 Not all questions are created equal. RAG techniques that do well on simple questions will fail on complex multi-step ones; fancy agentic techniques that address the latter will be way slower for simple questions. Adaptive RAG is a cool paper by Soyeong Jeong et al. that dynamically selects the best RAG strategy based on the query complexity. To do this, the authors propose a classifier that can classify different complexity levels. Thanks to Ravi Theja Desetty, we’re excited to implement the core idea as a Mistral AI cookbook, thanks for our Router abstraction. Thanks Sophia Yang, Ph.D. for the reviews! Check out our LlamaPack: https://lnkd.in/g3XQmfDb Paper: https://lnkd.in/d6DbStSK
-
-
Dive into the 🌪️ LlamaIndex + Mistral AI Cookbook Series! Your go-to guide for creating everything from simple to advanced RAG agents 🤖💡. Perfect for both beginners and pros: 1️⃣ RAG Setup 2️⃣ Efficient Routing 3️⃣ Smart Query Decomposition 4️⃣ Advanced Agents & Tools 🛠️ 5️⃣ Explore Adaptive RAG (🆕 research by Soyeong Jeong et al.) I'm giving this cookbook a try! 🧑🍳❤️ #AI #RAG #AgentBuilding #MistralAI
LlamaIndex + Mistral AI Cookbook Series 🧑🍳❤️ Here’s a definitive set of cookbooks to build simple-to-advanced RAG, agentic RAG, and agents in general with Mistral AI. It takes you through a tour of our RAG abstractions (including routing and query decomposition), along with our FunctionCallingAgent and ReActAgent. 1️⃣ RAG setup 2️⃣ Routing 3️⃣ Sub-question query decomposition 4️⃣ Agents + Tool Use with native function calling support 5️⃣ Adaptive RAG (new paper by Soyeong Jeong et al.) Huge shoutout to Ravi Theja Desetty for the work on this. Check out all the notebooks here: https://lnkd.in/gxTzy9jV
-
-
LlamaIndex + Mistral AI Cookbook Series 🧑🍳❤️ Here’s a definitive set of cookbooks to build simple-to-advanced RAG, agentic RAG, and agents in general with Mistral AI. It takes you through a tour of our RAG abstractions (including routing and query decomposition), along with our FunctionCallingAgent and ReActAgent. 1️⃣ RAG setup 2️⃣ Routing 3️⃣ Sub-question query decomposition 4️⃣ Agents + Tool Use with native function calling support 5️⃣ Adaptive RAG (new paper by Soyeong Jeong et al.) Huge shoutout to Ravi Theja Desetty for the work on this. Check out all the notebooks here: https://lnkd.in/gxTzy9jV
-
Tech Innovator & Strategic Thinker: Designing Future-Ready Solutions for Diverse Challenges
1moRobert Caulk it might interest you.