Craig Pfeifer

In this Technical Deep Dive, we share our approaches in using Large Language Models (#LLMs) to make numerical-type data(sets) like DHS, JMP, HDX more accessible with the example of WASH data. In this 15 min. video Olivier Mills and team cover: 1️⃣ Basic scripting to extract core data from datasets from API endpoints 2️⃣ How to use LLMs to generate LLM-friendly schema definitions 3️⃣ How to structure a AI-friendly database store that can scale to thousands of datasets and millions of data indicators 4️⃣ How to use vector querying to retrieve relevant indicators from the store based off of a natural language (english) query 5️⃣ How to build a prompt that generates valid SQL query using LLMs with the above retrieved data schema. The various components form a single AI system combining data ingestion and data exploration using a software layer, a data layer and multiple LLM layers. ⏭️ Next steps will be to expand the datasets, automate other data source ingestion and further test the system with more complex plain english queries. We plan to incorporate this as a new feature for WASH AI after testing. Feel free to reach out if you would like a copy of the Notebooks, sample prompts and data schema definitions. We are #opensource The initial work was done in collaboration with Data for Good YYC (see https://lnkd.in/g3AijS93)

3

1 Comment

Are you interested in vector search, indexing, or database efficiency? Then you won't want to miss this talk by Etienne Dilocker from Weaviate at Carnegie Mellon University's Database Group with Andy Pavlo! Etienne breaks down key challenges and innovations that are powering the future of search: ✨ Bit-Sliced Indexing - tackling complex range queries like price filters with innovative indexing. ✨ BM25 Hybrid Search - combining keyword and vector search to handle nuanced queries. ✨ Graph-Based Indexing (HNSW) - optimizing for high update rates and recalls without latency drops. ✨ Overcoming the Minority Filter Problem - enhancing search relevance through in-place filtering, based on the ACORN paper. ✨ Weaviate's Roadmap - performance, enterprise readiness, and pioneering "Recommendations as a Service." Check out the full talk for an in-depth look at how Weaviate is setting new standards in search technology: https://lnkd.in/gavAZRrn

33

2 Comments

📝12 Best Practices for Research Data Sharing: 3. Data repositories enable sharing of research outputs in a FAIR way, including support for metadata quality and completeness. Learn more today ⬇️ tiny.utk.edu/Idkml

2

Current benchmarking practices for AI agents may lead to misguided investments that don't perform as promised in real-world applications. Understanding evaluation shortcomings is crucial for making informed decisions about AI adoption and implementation. https://zurl.co/5yVs

1

Benchmarking Federated Learning for Large Language Models with FedLLM-Bench Quick read: https://lnkd.in/g2qf5YBp Paper: https://lnkd.in/gV4JCdYG

26

WorFBench: A Benchmark for Evaluating Complex Workflow Generation in Large Language Model Agents Large Language Models (LLMs) have shown remarkable potential in solving complex real-world problems, from function calls to embodied planning and code generation. A critical capability for LLM agents is decomposing complex problems into executable subtasks through workflows, which serve as intermediate states to improve debugging and interpretability. While workflows provide prior knowledge to prevent hallucinations, current evaluation benchmarks for workflow generation face significant challenges. Read the full article: https://lnkd.in/eqv7pQCt Paper: https://lnkd.in/eWDxh6Vz

5

Webinar alert - April 30, 1PM EDT. George Anadiotis from LinkedDataOrchestration and George Demarest of #Aerospike will guide you through selecting the right graph database and essential selection criteria tailored to your organizational needs. https://lnkd.in/eDV-QQuB #GraphDatabase #DataManagement

10

📚 Master citations in #QuartoPub with #CiteDrive! Effortlessly manage references, customize styles, and more. 🔍💡 👉 Learn how: https://buff.ly/3zAQjCs #RMarkdown #RStats #DataScience #AcademicWriting #BibTeX #Quarto #Research

1

Text embeddings have become crucial for transforming textual data into dense vector representations (with context), enabling efficient processing across various modalities like text, images, and audio. This article provides a good summary on some of these (not all) embeddings available for text https://lnkd.in/gKp4ta9a 1. Gensim Word Embeddings (Free) Why Use It: Specializes in NLP with open-source models like Word2Vec and FastText, promoting transparency and ease of custom training. 2. Facebook Embeddings (Free) Why Use It: Trained on extensive corpora and supports over 100 languages, making it versatile for global applications. 3. AllenNLP Embeddings (Free) Why Use It: Provides NLP-focused embeddings like BERT and ELMo, with tools for fine-tuning and visualization. 4. OpenAI Embeddings (Paid) Why Use It: Comprehensive training on massive datasets allows for effective semantic understanding, making it suitable for advanced NLP tasks. Ideal for tasks requiring zero-shot learning. 5. MultiLingual BERT (Free) Why Use It: Handles multilingual datasets effectively, offering robust support for multiple languages. Perfect for applications requiring consistent performance across different languages, such as global customer support systems. 6. DistilBERT Base Uncased (Free) Why Use It: Smaller and faster than BERT, maintaining near-comparable performance with reduced resource usage. 7. MPNet V2 (Free) Why Use It: Excels in text similarity tasks with a Siamese architecture, providing accurate similarity measurements. 8. HuggingFace Embeddings (Free/Paid) Why Use It: Offers versatility with a wide range of embeddings (text, image, audio) and the ability to fine-tune on custom data. HF also provides wrappers to almost all open sourced models. 9. Longformer Base 4096 (Free) Why Use It: Supports long texts up to 4096 tokens, making it unique in handling lengthy documents without truncation. 10. RoBERTa (2022) (Free) Why Use It: Offers updated text representations with comprehensive training on recent data. 11. GTE-Base (Free) Why Use It: Offers a good balance of performance and resource requirements with a 768-dimensional embedding. 12. GTE-Large (Free) Why Use It: Provides high-quality embeddings with more detailed representations, useful for nuanced text analysis. Best for applications requiring in-depth text understanding, such as detailed content recommendations. 13. GTE-Small (Free) Why Use It: Optimized for speed with a smaller model size, making it efficient for applications needing quick processing. Useful for real-time systems where fast text similarity searches are critical. 14. E5-Small (Free) Why Use It: Combines compact size with good performance, making it suitable for fast, general-purpose text embedding tasks.

11

Unlock the power of AI for research and reporting. This article by Robert Martin-Short shows how to build an agent using LangGraph and Tavily that not only extracts and organizes information but also autonomously generates and revises reports. #LLM #LangGraph

5

SATURDAY SCIENCE https://lnkd.in/guaNXeZT "Machine-learning models can fail when they try to make predictions for individuals who were underrepresented in the datasets they were trained on. For instance, a model that predicts the best treatment option for someone with a chronic disease may be trained using a dataset that contains mostly male patients. That model might make incorrect predictions for female patients when deployed in a hospital. To improve outcomes, engineers can try balancing the training dataset by removing data points until all subgroups are represented equally. While dataset balancing is promising, it often requires removing large amount of data, hurting the model’s overall performance. MIT researchers developed a new technique that identifies and removes specific points in a training dataset that contribute most to a model’s failures on minority subgroups. By removing far fewer datapoints than other approaches, this technique maintains the overall accuracy of the model while improving its performance regarding underrepresented groups."

3

Using deterministic NLP models in conjunction with Phi-3 to summarize long form documents, speeding up the review process and improving accuracy. #NLP #Phi3 #LLM

2

🚀 Judge0 facilitates AI research! 🎓 Carnegie Mellon University researchers recently explored an intriguing question in their paper, "ECCO: Can We Improve Model-Generated Code Efficiency Without Sacrificing Functional Correctness?". To tackle this challenge, they heavily relied on Judge0. 👉 "To perform reliable and reproducible executions, we introduce an evaluation setup using a cloud-hosted code execution engine, Judge0, which produces stable execution output on correctness, runtime, and memory usage, thanks to its agnostic nature to local hardware specifications. It supports up to 66 programming languages (PLs), allowing future work to extend to other languages." - Waghjale, Veerendranath et al. 📖 Read the full paper here: https://lnkd.in/ebqFt-NY #AI #Judge0 #LLM #CodeExecution #CodeEfficiency #ArtificialIntelligence #Coding #Programming

9

📚 New Research: Enhancing Reasoning in Large Language Models with RE2 🧠 This recent study introduces a simple yet effective prompting strategy called RE2 (Re-Reading) to improve reasoning capabilities in Large Language Models (LLMs). Unlike traditional approaches that focus on eliciting reasoning through outputs, RE2 emphasizes understanding by re-processing the input twice. This method enables LLMs to achieve a "bidirectional" understanding, which is particularly beneficial in unidirectional decoder-only models like GPT-3 and LLaMA. Through extensive testing across 14 datasets and 112 experiments, RE2 consistently enhanced performance in reasoning tasks, demonstrating its adaptability across various models and settings. Whether used in zero-shot, few-shot, or in combination with thought-eliciting strategies like Chain-of-Thought, RE2 shows promise in making LLMs more reliable and versatile in complex problem-solving. 🔗 #AI #MachineLearning #NaturalLanguageProcessing #Reasoning #LLM

1

Hallucinations in LLMs occur when the models generate text that appears plausible but is factually incorrect or nonsensical. Several strategies can help mitigate these hallucinations. https://lnkd.in/g_uFvVfX

1

Enhancing Self-Supervised Learning with Automatic Data Curation: A Hierarchical K-Means Approach Quick read: https://lnkd.in/giitcGQy Paper: https://lnkd.in/gyN-2sWa

18