Post Content
Enhance your AI models using advanced fine‑tuning and distillation techniques that deliver higher accuracy and efficiency. This session explores the latest techniques in Direct Preference Optimization (DPO), Reward Fine-Tuning (RFT), and model distillation within Azure OpenAI. Learn to optimize performance while reducing data requirements and operational costs for smoother deployments.
To learn more, please check out these resources:
* https://aka.ms/build25/plan/CreateAgenticAISolutions
𝗦𝗽𝗲𝗮𝗸𝗲𝗿𝘀:
* Srinivas Gadde
* Omkar More
* Alicia Frame
𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗜𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻:
This is one of many sessions from the Microsoft Build 2025 event. View even more sessions on-demand and learn about Microsoft Build at https://build.microsoft.com
BRK150 | English (US) | AI, Copilot & Agents
#MSBuild
Chapters:
0:00 – Overview of Today’s Session Topics
00:16:01 – Explanation of Model Evaluation and Deploy Process
00:19:15 – Practical Demonstration of Model Fine-Tuning and Deployment
00:30:32 – Demonstration of Reinforcement Fine Tuning with Limited Data Samples
00:34:19 – Completion of Reinforcement Fine Tuning and Deployment
00:47:46 – Benefits of Fine Tuning for Latency
00:49:25 – Model Choices and Fine Tuning Versus Prompt Engineering Read More Microsoft Developer
