Why Vision Language Models Ignore What They See with Munawar Hayat - #758

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 contrastive learning designed to solve complex, composed retrieval tasks—such as searching via combined text and image queries—without increasing inference costs. Finally, we cover the difficulties generative models face when rendering multiple human subjects, and the new "MultiHuman Testbench" his team created to measure and mitigate issues like identity leakage and attribute blending. Throughout the discussion, we examine how these innovations align with the need for efficient, on-device AI deployment. The complete show notes for this episode can be found at

You can listen to Why Vision Language Models Ignore What They See with Munawar Hayat - #758 online on Radio and Podcast. Open the player on this page to stream the available audio.

Why Vision Language Models Ignore What They See with Munawar Hayat - #758 is an episode from The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) by TWIML.

This episode is 57:40 long.

This episode was published on Dec 9, 2025.

Yes. Use the heart button on the episode page to add it to your favorite episodes list.

Yes. This page shows related episodes from The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) when more episodes are available from the podcast feed.