Nouvelles
Chargement
![Research Focus: Week of April 29, 2024](https://www.microsoft.com/en-us/research/uploads/prodnew/2024/04/RF40-BlogHeroFeature-1400x788-1-480x280.png)
Microsoft Research Blog
Research Focus: Week of April 29, 2024
In this edition: Can LLMs transform natural language into formal method postconditions; Semantically aligned question + code generation for automated insight generation; Explaining CLIP performance disparities on blind/low vision data; plus recent news.
![A graphic overview of the way performance assessment methods change across the development lifecycle. It has four phases: getting started, connecting with users, tuning the user experience, and performance assessment in the deployment context. It visually shows how the balance of user experience and tech development change over these four phases.](https://www.microsoft.com/en-us/research/uploads/prod/2022/09/1400x788_Project_Tokyo_herov3-480x280.jpg)
Microsoft Research Blog
Assessing AI system performance: thinking beyond models to deployment contexts
| Cecily Morrison, Martin Grayson, and Camilla Longden
AI systems are becoming increasingly complex as we move from visionary research to deployable technologies such as self-driving cars, clinical predictive models, and novel accessibility devices. Unlike singular AI models, it is more difficult to assess whether these more complex…