Leveraging LLMs for Phishing Detection and Takedowns

In this webinar, we discuss how we have trained and deployed state-of-the-art large language models from the BERT and GPT families in the Bolster Security platform to elevate the efficacy of protection we deliver to our customers. These models plug into our infrastructure on multiple levels, and fully automate our workflows from detection and monitoring, through to takedowns. Our host will take you through the performance of these large language models on standardized metrics. Furthermore, we will expand on the ideas of strategies to integrate LLMs into software pipelines like zero-shot and fine-tuning based learning. In this webinar, you’ll hear from the Bolster threat research team on: - Our revamped intent detection module, which now integrates Convolutional BERT LLM using the revolutionary Transformer architecture. - How Bolster has simplified the training and testing of our models with HuggingFace’s Transformer API. - The explosion on the Natural Language Processing landscape caused by Generative AI based LLMs like GPT and LLaMa and how Bolster is harnessing this technology. - The all-new automated takedowns for tackling fake mobile app phishing attacks, which deploys a state-of-the-art generative AI-based LLM to eliminate the need of any human intervention in takedowns. Join us Tuesday, October 17th at 11am PST/2pm EST