This AI Paper Identifies Function Vector Heads as Key Drivers of In-Context Learning in Large Language Models

In-context learning (ICL) is something that allows large language models (LLMs) to generalize & adapt to new tasks with minimal demonstrations. ICL is crucial for improving model flexibility, efficiency, and…

Project Alexandria: Democratizing Scientific Knowledge Through Structured Fact Extraction with LLMs

Scientific publishing has expanded significantly in recent decades, yet access to crucial research remains restricted for many, particularly in developing countries, independent researchers, and small academic institutions. The rising costs…

Step by Step Guide to Build an AI Research Assistant with Hugging Face SmolAgents: Automating Web Search and Article Summarization Using LLM-Powered Autonomous Agents

Hugging Face’s SmolAgents framework provides a lightweight and efficient way to build AI agents that leverage tools like web search and code execution. In this tutorial, we demonstrate how to…

Agentic AI vs. AI Agents: A Technical Deep Dive

Artificial intelligence has evolved from simple rule-based systems into sophisticated, autonomous entities that perform complex tasks. Two terms that often emerge in this context are AI Agents and Agentic AI.…

Rethinking MoE Architectures: A Measured Look at the Chain-of-Experts Approach

Large language models have significantly advanced our understanding of artificial intelligence, yet scaling these models efficiently remains challenging. Traditional Mixture-of-Experts (MoE) architectures activate only a subset of experts per token…

Accelerating AI: How Distilled Reasoners Scale Inference Compute for Faster, Smarter LLMs

Improving how large language models (LLMs) handle complex reasoning tasks while keeping computational costs low is a challenge. Generating multiple reasoning steps and selecting the best answer increases accuracy, but…

Defog AI Open Sources Introspect: MIT-Licensed Deep-Research for Your Internal Data

Modern enterprises face a myriad of challenges when it comes to internal data research. Data today is scattered across various sources—spreadsheets, databases, PDFs, and even online platforms—making it difficult to…

HippoRAG 2: Advancing Long-Term Memory and Contextual Retrieval in Large Language Models

LLMs face challenges in continual learning due to the limitations of parametric knowledge retention, leading to the widespread adoption of RAG as a solution. RAG enables models to access new…

NeoBERT: Modernizing Encoder Models for Enhanced Language Understanding

Encoder models like BERT and RoBERTa have long been cornerstones of natural language processing (NLP), powering tasks such as text classification, retrieval, and toxicity detection. However, while decoder-based large language…

Building a Collaborative AI Workflow: Multi-Agent Summarization with CrewAI, crewai-tools, and Hugging Face Transformers

CrewAI is an open-source framework for orchestrating autonomous AI agents in a team. It allows you to create an AI “crew” where each agent has a specific role and goal…