Janvi
Palan
Previously at Samsung Research America
"Building and evaluating robust conversational interfaces using LLMs"
"Building and evaluating robust conversational interfaces using LLMs"
24 - 12:50 PM EDT/New York (See in local time)
Add to Calendar
05/24/2025 12:50 PM
05/24/2025 01:10 PM
America/New_York
#WTGC2025
"Building and evaluating robust conversational interfaces using LLMs"
#WTGC2025
"Building and evaluating robust conversational interfaces using LLMs"
https://www.womentech.net/ringcentral
https://www.womentech.net/ringcentral
"Building and evaluating robust conversational interfaces using LLMs"
"Building and evaluating robust conversational interfaces using LLMs"
Don’t miss out and join visionaries, innovators, and thought leaders from all over the world at the Women in Tech Global Conference.
Vote by Sharing
Unite 100 000 Women in Tech to Drive Change with Purpose and Impact.
Do you want to see this session? Help increase the sharing count and the session visibility. Sessions with +10 votes will be available to career ticket holders.
Please note that it might take some time until your share & vote is reflected.
Session: Building and evaluating robust conversational interfaces using LLMs
The rapid advancements in machine learning and natural language processing have led to a surge in the development of conversational agents. In this talk, I present an innovative framework for designing, evaluating, and deploying human-like conversational agents in retail and hospitality. The framework addresses challenges such as conversation rigidity, tonality, and latency in large language models. Additionally, it incorporates customer-data pipelines for personalized interactions and robust conversational interfaces.We will demonstrate the creation of a performance baseline using off-the-shelf models, evaluate the retrieval pipeline, and how to bootstrap annotated datasets for fine-tuning task-specific models. This cutting-edge framework paves the way for next-generation conversational agents, revolutionizing customer interactions across industries.Key Takeaways
- Generative AI
- Large Language Models
- Natural Language Understanding
- AI ML
Bio
Janvi is a Senior ML research engineer at Samsung Research America. Previously working on Samsung's artificial humans division, she lead the realistic behavior and conversation generation component of virtual humans - while ensuring that the quality and latency of interaction with an artificial human is not affected. She works on a variety of AI ML related projects, specially focused in the LLM, search and retrieval space. She loves taking vague, undefined problems and translating them into tangible pieces that can be tackled.