Munawar Hayat
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
Why Vision Language Models Ignore What They See with Munawar Hayat
- Published
- December 9, 2025
- Duration
- 57:40
- Summary source
- description
- Last updated
- Apr 21, 2026
Discusses multimodal, generative-ai.
Summary
In this episode, we’re joined by Munawar Hayat, researcher at Qualcomm AI Research, to discuss a series of papers presented at NeurIPS 2025 focusing on multimodal and generative AI. We dive into the persistent challenge of object hallucination in Vision-Language Models (VLMs), why models often discard visual information in favor of pre-trained language pr…
Show notes
In this episode, we’re joined by Munawar Hayat, researcher at Qualcomm AI Research, to discuss a series of papers presented at NeurIPS 2025 focusing on multimodal and generative AI. We dive into the persistent challenge of object hallucination in Vision-Language Models (VLMs), why models often discard visual information in favor of pre-trained language priors, and how his team used attention-guided alignment to enforce better visual grounding. We also explore a novel approach to generalized cont
Themes
- multimodal
- generative-ai