Skip to main content
Webinar
Thu, Apr 24, 2:00 PM - 3:30 PM (UTC)

Detecting and Preventing LLM Hallucinations

About this event

This event is free for community members. If you want to participate in the event or get the recording, you have to firstly register in the community (see instructions below) and then click on Attend to reserve your spot. All registrants will get a webinar recording after the event and a certificate.

For many LLM use cases, quality means that you have hallucinations under control.  Join this webinar to explore methods for detecting and mitigating hallucinations so that you can confidently go live with LLMs.

Attendees will learn

  • How to evaluate LLM outputs
  • Hallucination detection strategies
    • Confidence scoring
    • Consistency checking
    • External knowledge verification
    • Citation verification
    • Adversarial prompt testing
  • Hallucination prevention strategies such as
    • Firewall guardrails
    • Retrieval-augmented generation (RAG)
    • Prompt optimization
    • Fine Tuning
    • Reinforcement Learning
    • Citations

About James: 

James Massa is a global keynote speaker and serial innovator with 6 patents for LLM, automated testing, MLOps, FinOps and data quality. He is a senior executive at JPMorganChase who won the 2024 FSTech award for leading the Best Financial Services IT Team. James holds master’s degrees in digital design from Harvard University and in finance from Baruch.

 How to register for an event, if you are not a member:

 

  1. Create an account.

Password advice: use the password generator for passwords or avoid dicitionary words ( even when using special characters).

  1. Check your email box, there is an email from ShiftSync. Click on a button in the email and activate your account. 
  2. Go back to this page and click Attend.
  3. Now you are registered! ✨

 

Attendees (16)
musayusuf
Apprentice
DamiS
Space Cadet
troopesh
Ensign
Jeethu
Ensign
Pratik mohanty
Space Cadet
sarika77
Ensign
Event details
Online event
Thu, Apr 24, 2:00 PM - 3:30 PM (UTC)
Add to Calendar