90/30 Club Reading: Reasoning-Driven Synthetic Data Generation and Evaluation

​Come join us for a group discussion of "Reasoning-Driven Synthetic Data Generation and Evaluation"

Paper link: https://arxiv.org/abs/2603.29791v1

Blog post: https://research.google/blog/designing-synthetic-datasets-for-the-real-world-mechanism-design-and-reasoning-from-first-principles/

​LLMs can only generate the average text, right? Right?

To address the scarcity of data required for specialized AI, we introduce Simula, a framework that reframes synthetic data generation as dataset-level mechanism design. By using reasoning to architect datasets from first principles, Simula enables fine-grained control over coverage, complexity, and quality, providing scalable generation for privacy-sensitive or data-scarce domains.

Bonus Round: Papers I wish I could get you to read #001: tracr

​Event Schedule:

​7pm-8pm: Quiet reading time, grab a snack and read! (optional)

​8pm-9pm: Group discussion about the paper 📝

​9pm-10pm: We have our space for a bit longer, stay to socialize or network!

​Our event is hosted within Mox SF, the gracious donors of the space where we will meet.