# Jul 5, 2024

## Beating NumPy matrix multiplication in 150 lines of C

A custom C implementation of matrix multiplication outperforms NumPy's performance by achieving over **1 TFLOPS** on an AMD Ryzen 7700 CPU, utilizing a design inspired by the [BLIS](https://en.wikipedia.org/wiki/BLIS_(software)) framework and optimized through parallelization with OpenMP.

## License Update

Stability AI introduces a **new Community License** allowing free use of its models for research, non-commercial, and commercial purposes, provided annual revenues do not exceed USD$1M.

## Likelihood computation in diffusion models [P]

**Diffusion models**, specifically the **SDE approach** detailed in [Song et al.](http://arxiv.org/abs/2011.13456) and [Song et al.](http://arxiv.org/abs/2101.09258), require the **probability flow ODE** for exact likelihood computation of generated data, due to the intractability of SDE likelihoods.

## [R] Taxonomy for Data Transformations in AI Systems

The **Data Transformation Taxonomy** introduced in the SIGMOD'24 paper categorizes **data transformations in AI systems** into reusable across models, model-specific, and real-time request-specific. [Read the SIGMOD paper](https://dl.acm.org/doi/10.1145/3626246.3653389)

## [N] Moshi very first voice-enabled AI openly accessible to all

**Moshi**, the **first voice-enabled AI** openly accessible to all, offers **natural conversation** capabilities with a **low latency**, currently limited to **5 minutes** and available in **English**. [Keynote video](https://www.youtube.com/live/hm2IJSKcYvo?si=7CJ8Mw7O8H0xOwVq&t=418) and [Press release](https://kyutai.org/cp_moshi.pdf)

## [R] Problems with Grad-CAM visualization for medical imaging

**Grad-CAM visualizations** in medical imaging, specifically for **grayscale chest thermograms**, often **highlight irrelevant features** such as arms and shoulders, **overlooking the chest area** crucial for detecting abnormalities.

## [D] Constrained decoding as stateful navigation?

**Constraining a Large Language Model's (LLM's) decoder to specific grammars** like XML or JSON can enhance the reliability of a wrapping program's parsing process, but may be suboptimal without grammar-specific training.

## TwoMinutePapers - ChatGPT Just Learned To Fix Itself!

**OpenAI's new paper** introduces a groundbreaking concept where **AI systems critique and correct errors** in other AI-generated outputs, enhancing code quality and security.