Machine Learning
275 articles in this category (Page 12 of 12)
7 Advanced Feature Engineering Tricks for Text Data Using LLM Embeddings
Explore seven advanced techniques to enhance text-based machine learning models by combining LLM-generated embeddings with traditional features, improving accuracy in tasks like sentiment analysis and clustering.
Microsoft Releases Agent Lightning: A Reinforcement Learning Framework for Optimizing AI Agents
Microsoft introduces Agent Lightning, an open-source framework that enables reinforcement learning (RL)-based training of large language models (LLMs) for AI agents without requiring changes to existing agent stacks.
Liquid AI Releases LFM2-ColBERT-350M: A Compact Late Interaction Model for Multilingual Cross-Lingual Retrieval
Liquid AI introduces LFM2-ColBERT-350M, a 350M-parameter late interaction retriever optimized for multilingual and cross-lingual search, offering high accuracy and fast inference speeds.
Vector Sync Patterns: Keeping AI Features Fresh When Your Data Changes
Ricardo Ferreira shares 5 essential Vector Sync Patterns designed to solve the complex, multi-dimensional challenges of vector staleness and synchronization in AI-driven microservices. He explains how to leverage event-driven architecture (Kafka/Flink) and CDC to reliably manage data, application, and business-driven changes for architects and senior developers.
Google Open-Sources Coral NPU Platform for AI on Edge Devices
Google Research has open-sourced the Coral NPU platform to facilitate the integration of AI into wearables and edge devices, addressing challenges related to performance, fragmentation, and user privacy.
Introducing Thinking-in-Modalities with TerraMind: A Novel Approach to Foundation Models
IBM Research introduces TerraMind, a foundation model leveraging 'Thinking in Modalities' (TiM) to enhance Earth observation analysis by generating missing data representations.
Hugging Face Enhances Dataset Streaming for 100x Efficiency
Hugging Face has significantly improved dataset streaming capabilities in their 'datasets' and 'huggingface_hub' libraries, enabling faster and more efficient training on large datasets. Key improvements include reduced API requests, faster data resolution, and enhanced control over streaming pipelines.