Introduction: The Need for Efficient RL in LRMs Reinforcement Learning RL is increasingly used to enhance LLMs, especially for reasoning tasks. These models, known as Large Reasoning Models (LRMs), generate intermediate “thinking” steps before providing final answers, thereby improving performance…
The Challenge of Fine-Tuning Large Transformer Models Self-attention enables transformer models to capture long-range dependencies in text, which is crucial for comprehending complex language patterns. These models work efficiently with massive datasets and achieve remarkable performance without needing task-specific structures.…
In this tutorial, we delve into building an advanced data analytics pipeline using Polars, a lightning-fast DataFrame library designed for optimal performance and scalability. Our goal is to demonstrate how we can utilize Polars’ lazy evaluation, complex expressions, window functions,…
Research has shown that large language models (LLMs) tend to overemphasize information at the beginning and end of a document or conversation, while neglecting the middle. This “position bias” means that, if a lawyer is using an LLM-powered virtual assistant…
MIT Morningside Academy for Design (MAD) Fellow Caitlin Morris is an architect, artist, researcher, and educator who has studied psychology and used online learning tools to teach herself coding and other skills. She’s a soft-spoken observer, with a keen interest in how…
During his first year at MIT in 2021, Matthew Caren ’25 received an intriguing email inviting students to apply to become members of the MIT Schwarzman College of Computing’s (SCC) Undergraduate Advisory Group (UAG). He immediately shot off an application. Caren…
The Challenge of Updating LLM Knowledge LLMs have shown outstanding performance for various tasks through extensive pre-training on vast datasets. However, these models frequently generate outdated or inaccurate information and can reflect biases during deployment, so their knowledge needs to…
Python A2A is an implementation of Google’s Agent-to-Agent (A2A) protocol, which enables AI agents to communicate with each other using a shared, standardized format—eliminating the need for custom integration between services. In this tutorial, we’ll use the decorator-based approach provided…
On May 6, MIT AgeLab’s Advanced Vehicle Technology (AVT) Consortium, part of the MIT Center for Transportation and Logistics, celebrated 10 years of its global academic-industry collaboration. AVT was founded with the aim of developing new data that contribute to…
The Need for Efficient On-Device Language Models Large language models have become integral to AI systems, enabling tasks like multilingual translation, virtual assistance, and automated reasoning through transformer-based architectures. While highly capable, these models are typically large, requiring powerful cloud…
