About
Research & Development
Assessing LLM Performance in Python Type Inference: Introducing TYPY BENCH
TLDR: TYPY BENCH is a new benchmark for evaluating how well Large Language Models (LLMs) can infer types in untyped Python codebases. It introduces...
Research & Development
How Persona Prompts Influence Large Language Model Security
TLDR: This research explores how "persona prompts" can be used to bypass Large Language Model (LLM) safety mechanisms, making them generate harmful content. The...
Research & Development
Measuring Time-Series Data Relationships with Wasserstein Distance for Foundation Models
TLDR: A new research paper proposes using Wasserstein distance to measure the similarity between time-series datasets by modeling them as multivariate normal distributions. This...
Research & Development
Enhancing Smart Grid Security with AI: A Hybrid Approach for Explainable Cyberattack Detection
TLDR: This research introduces a hybrid framework that combines machine learning (ML) for fast detection of cyberattacks in smart grids with Large Language Models...
Research & Development
Optimizing LLM Specialization: A New Framework for Data Source Evaluation
TLDR: This research introduces a framework for optimizing domain-specific dataset construction in foundation model training. It proposes using 'scaling laws' instead of single 'point...
Research & Development
AI-Powered Simulations: A New Approach to Online Safety Education
TLDR: A research paper introduces a system that uses two Large Language Models (LLMs) to simulate unsafe online conversations, specifically scams. One LLM acts...