Explore this 36-minute talk from the University of Central Florida that investigates whether language models can be influenced to perceive and interpret the world in alternative ways. The presentation examines the malleability of AI models' worldviews through verbal interaction and discusses the implications of being able to shift how these systems understand and respond to information.
Overview
Syllabus
Paper 6 - CAN WE TALK MODELS INTO SEEING THE WORLD DIFFERENTLY?
Taught by
UCF CRCV
Tags
Related Courses
-
AI & Cognitive Science: Bridging Minds and Machines
-
Visual-Language Models Introduction - Part 2 - Lecture 6
-
MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models
-
Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding
-
From Images to Text: New Forms of Human-AI Interaction
-
Models Within Models - How Do LLMs Represent the World?
