Kanika Narang

Meta will be at The AI Conference this fall!
Research Scientist
Meta Platforms Inc.

Presentation Title:

On-Device LLMs: Unleashing the Power of Conversational AI for Mobile and Wearable Devices

Presentation Summary:

Atom icon for The AI Conference 2023, a groundbreaking two-day event on AGI, LLMs, Infrastructure, Alignment, AI Startups, and Neural Architectures.Large Language Models (LLMs) have captured the attention of the tech world with their remarkable common-sense reasoning and generalizability. However, their large size and server transfer requirements can make them resource-intensive and slow, which is problematic for use in mobile or wearable devices like smart glasses and smart watches. Moreover, on-device computing could offer a solution to privacy concerns by keeping sensitive data, such as text messages or photos, on the device itself. To tackle these challenges, we’ve developed a more compact language model, ranging from 0.5B to 1.4B parameters. This model is designed to run on-device, providing a competitive performance for conversational grounded tasks, while also managing latency and memory usage effectively. 

Brain icon for The AI Conference 2023, a groundbreaking two-day event on AGI, LLMs, Infrastructure, Alignment, AI Startups, and Neural Architectures.

In this presentation, I’ll delve into our work on creating a versatile, on-device LLM, which is a distilled version of the LLAMA model, specifically tailored for conversational reasoning tasks. I’ll discuss our pretraining framework and our innovative approach to finetuning, which involves using LLM synthesized, task-specific data in a fresh dialogue format. Finally, I’ll share our strategy for scaling our text-based conversational model to a multimodal model, enhancing generative experiences such as composing text replies, document summarization, image captioning, and visual question-answering on wearables or mobile devices.

About | Kanika Narang

Kanika is a seasoned professional with a decade of practical experience in the field of AI, focusing on research, development, and implementation of AI models to address various complex issues, such as content safety, user behavior modeling, and recommendation systems. In her current role as a Research Scientist at Meta Reality Labs, she is involved in pioneering conversational AI projects for AR devices. Kanika earned her PhD from the University of Illinois, Urbana-Champaign (UIUC), where her research delved in the intersection of Deep Learning, Computational Social Science, and Graph Learning. She has contributed around 20 papers presented at esteemed AI/ML conferences like ACL, SIGIR, and ICML. Moreover, she has an extensive experience of serving as a speaker and panelist at numerous global conferences.