The Israeli Natural Language Processing Meetup (NLP IL)

NOTE: Due to the war, the original event was postponed.
Please re-register for this event, even if you RSVP'd before.
Also note the change in location.

Agenda:

18:00-18:30 - Gathering, food, and drinks
18:30-18:45 - NLP IL & Hyro : Opening words
18:45-19:15 - Nimrod Sagie (Hyro)
AI-Powered IVRs: Real-World Lessons
19:15-19:45 - Danny Lazarev (WIZ) -
Enhancing Secret Detection with SLMs
19:45-20:15 - Mike Erlihson, PhD (Stealth) -
Making Text Generation With LLMs Faster

Abstracts:

Lecture #1 - Nimrod Sagie (Hyro) - AI-Powered IVRs: Real-World Lessons
Deploying AI-powered call center agents holds the promise of transforming the caller experience, but reaching production is far from straightforward. In this talk, we’ll walk through our hands-on journey building and launching agents for a healthcare call center. You’ll hear how we tackled everything from natural language intent mapping and advanced call routing, to resolving ambiguity and addressing the quirks of large language models. We’ll dive into real-world failures, unexpected challenges, and the practical decisions that helped us create a system that’s both scalable and reliable.

Lecture #2 - Danny Lazarev (WIZ) - Enhancing Secret Detection with SLMs
Dive into the challenges of LLMs in cybersecurity as we explore the process of fine-tuning a Small Language Model to handle the task of secret detection in code and be efficient enough to run on any laptop.

Lectrue #3 - Mike Erlihson, PhD (Stealth) - Making Text Generation With LLMs Faster
In this lecture we will explore different techniques to accelerate text generation for self-hosted LLMs.

For the next meeting of the Vision-Language Club we are very excited to host Lior Cohen for a talk on Vision-Language Models at NVIDIA: Advancing Multimodal Agents and Physical AI.

We will meet online, on the 23rd of July at 20:00. The talk will also be recorded and uploaded to NLP-IL's YouTube channel. The talk will be conducted in Hebrew.

This session explores the real-world impact of vision-language models (VLMs) as they move beyond the research lab, showcasing their contributions to multimodal agentic workflows and the rapidly advancing field of Physical AI and robotics. We will explore NVIDIA’s latest solutions across these domains, through the lens of its full-stack approach for intelligent systems across diverse environments. The discussion will incorporate insights from relevant research papers and practical industrial perspectives.
Special attention will be given to Vision-Language-Action (VLA) models - a new class of foundation models designed to integrate perception, natural language understanding, and physical actions. We will review NVIDIA’s most recent research in this space, including the Gr00t N1 model.

Lior Cohen is a senior GenAI solution architect at NVIDIA, specializing in large multimodal models and AI agents. Lior completed her master’s degree studies at Ben Gurion University. She is also a member of the leading team of NLP IL community.