Better RAG using Relevant Information Gain

Hosted by Amir Feizpour and Marc Pickett

Fri, Sep 5, 2025

4:00 PM UTC (45 minutes)

Virtual (Zoom)

Free to join

Invite your network

Go deeper with a course

Build Multi-Agent Applications - A Bootcamp - LangGraph, Cursor, n8n
Amir Feizpour, PhD
View syllabus

What you'll learn

about balancing diversity and relevance in retrieved passage

how relevant information gain can optimize retrieval

how this approach achieves SOTA results on benchmarks

Why this topic matters

RAG systems boost LLM performance by adding retrieved context, but limited context windows make redundancy costly. This talk introduces a simple relevant information gain metric that naturally balances diversity and relevance, improving retrieval quality and achieving state-of-the-art QA results without complex trade-offs.

You'll learn from

Amir Feizpour

Founder @ Aggregate Intellect

Amir Feizpour is the founder, CEO, and Chief Scientist at Aggregate Intellect building a generative business brain for service and science based companies. Amir has built and grown a global community of 5000+ AI practitioners and researchers gathered around topics in AI research, engineering, product development, and responsibility. Prior to this, Amir was an NLP Product Lead at Royal Bank of Canada. Amir held a research position at University of Oxford conducting experiments on quantum computing resulting in high profile publications and patents. Amir holds a PhD in Physics from University of Toronto. Amir also serves the AI ecosystem as an advisor at MaRS Discovery District, works with several startups as fractional chief AI officer, and engages with a wide range of community audiences (business executives to hands-on developers) through training and educational programs. Amir leads Aggregate Intellect’s R&D via several academic collaborations.

Marc Pickett

AI Researcher @ Emergence AI

Marc has pursued AGI in various guises for over 30 years, starting with Reinforcement Learning with Satinder Singh and Andy Barto, then moving to unsupervised "Robot babies" with Tim Oates and David Aha, and finally working at Google AI on a team led by Ray Kurzweil, where Marc worked on Continual Learning. Since 2023, Marc has worked at a startup, Emergence AI, where he works on Long-term memory for automated assistants. His team recently achieved state of the art performance on the LongMemEval benchmark for evaluating long-term memory.

Sign up to join this lesson

By continuing, you agree to Maven's Terms and Privacy Policy.

© 2025 Maven Learning, Inc.