Image Decision Models for RPA: Forms, Scans and Screenshots

In this video, we return to looking at decision models, but this time for images, with the use case being RPA. 🤗 HF: https://huggingface.co/mohit67890/imajev-4b 🤗 HF: https://huggingface.co/akhilaaa3/Jev-Omni Sam's 2nd Channel: https://www.youtube.com/channel/UCUF7WtR4JpVbQqcQGrLeADw Twitter: https://x.com/Sam_Witteveen 🕵️ Interested in building LLM Agents? Fill out the form below Building LLM Agents Form: https://drp.li/dIMes 👨‍💻Github: https://github.com/samwit/llm-tutorials ⏱️Time Stamps: 00:00 Intro 00:17 RPA Automates Steps, Not Decisions 00:39 Why Custom Deep Learning Models Fell Short 00:59 The Form Inspector & Two Open Models 01:38 Why Focus on Images 01:56 The Problem: Checking Screenshots, Photos & Forms 02:42 RPA Story 03:24 What Went Wrong With RPA 05:05 ImageBench Leaderboard 05:24 The Story Behind ImaJev 07:29 Smart If Statements & Confidence Scores 08:01 Demo: Form Decision Inspector 09:11 Running ImaJev 4B vs Jev Omni 10:53 Choice vs True/False Questions 11:57 Follow-Up Emails With Conditional Logic 12:41 Adding New Questions 13:21 Testing a Second Form 14:35 Confidence Thresholds Per Question