Modern LLMs can now produce highly readable abstractive summaries, to the point where traditional automated metrics for evaluating summary quality, such as ROUGE, have become saturated. However, LLMs still sometimes introduce unwanted content into summaries, i.e., information inconsistent with or unsupported by their source. Measuring the occurrence of these often subtle ``hallucinations'' automatically has proved to be challenging. This in turn has motivated development of a variety of metrics intended to measure the factual consistency of generated summaries against their source. But are these approaches measuring what they purport to do? In this work, we stress-test automatic factuality metrics. Specifically, we investigate whether and to what degree superficial attributes of summary texts suffice to predict ``factuality'', finding that a (supervised) model using only such shallow features is reasonably competitive with SOTA factuality scoring methods. We then evaluate how factuality metrics respond to factual corrections in inconsistent summaries and find that only a few show meaningful improvements. In contrast, some metrics are more sensitive to benign, non-factual edits. Motivated by these insights, we show that one can ``game'' (most) automatic factuality metrics, i.e., reliably inflate ``factuality'' scores by appending innocuous sentences to generated summaries. Taken together, our results raise questions about the degree to which we should rely on existing automated factuality metrics and what exactly we want ``factuality metrics'' to measure.

Current machine learning methods for medical image analysis primarily focus on developing models tailored for their specific tasks, utilizing data within their target domain. These specialized models tend to be data-hungry and often exhibit limitations in generalizing to out-of-distribution samples. Recently, foundation models have been proposed, which combine data from various domains and demonstrate excellent generalization capabilities. Building upon this, this work introduces the incorporation of diverse medical image domains, including different imaging modalities like X-ray, MRI, CT, and ultrasound images, as well as various viewpoints such as axial, coronal, and sagittal views. We refer to this approach as multi-domain model and compare its performance to that of specialized models. Our findings underscore the superior generalization capabilities of multi-domain models, particularly in scenarios characterized by limited data availability and out-of-distribution, frequently encountered in healthcare applications. The integration of diverse data allows multi-domain models to utilize shared information across domains, enhancing the overall outcomes significantly. To illustrate, for organ recognition, multi-domain model can enhance accuracy by up to 10% compared to conventional specialized models.

Designing good data-driven models hugely depends on the quality of data. Well, data is a set of numbers, and shouldn't bother the developers much. As they say, the devil lies in the details, real data comes with a set of issues like imbalanced classes, inherent biases, unstructured values, etc. On the other hand, synthetic data provides the developers with the flexibility of scalability of data and freedom from biases, opening a whole lot of possibilities for creating a model that doesn't exist in the real world. In addition, synthetic data holds the benefits of protecting user data privacy all while giving the freedom to experiment with.

This article comes from Joon Im, a student in Business Science University. Joon has completed both the 201 (Advanced Machine Learning with H2O) and 102 (Shiny Web Applications) courses. Joon shows off his progress in this Web Scraping Tutorial with rvest. I recently completed the Part 2 of the Shiny Web Applications Course, DS4B 102-R and decided to make my own price prediction app. The app works by predicting prices on potential new bike models based on current existing data.

This article comes from Joon Im, a student in Business Science University. Joon has completed both the 201 (Advanced Machine Learning with H2O) and 102 (Shiny Web Applications) courses. Joon shows off his progress in this Web Scraping Tutorial with rvest. I recently completed the Part 2 of the Shiny Web Applications Course, DS4B 102-R and decided to make my own price prediction app. The app works by predicting prices on potential new bike models based on current existing data.

One of the reasons for the success of convolutional networks is their equivariance/invariance under translations. However, rotatable data such as molecules, living cells, everyday objects, or galaxies require processing with equivariance/invariance under rotations in cases where the rotation of the coordinate system does not affect the meaning of the data (e.g. object classification). On the other hand, estimation/processing of rotations is necessary in cases where rotations are important (e.g. motion estimation). There has been recent progress in methods and theory in all these regards. Here we provide an overview of existing methods, both for 2D and 3D rotations (and translations), and identify commonalities and links between them, in the hope that our insights will be useful for choosing and perfecting the methods.

The new MOBOTIX camera platform, jointly developed with Konica Minolta, is based on the distributed intelligence in our camera system and is crucial for artificial intelligence and at the same time the key for the communication of our products with other sensors and devices in the network to enable "Beyond Human Vision" solutions. The state-of-the-art analysis methods located on the camera itself helps users increase process efficiency and develop new business and revenue models. Future software updates will enable completely new functions based on deep learning methods, such as the recognition of human behavior, moods or voices. In order to make this possible, a so-called "plug-in concept" is being further developed that goes far beyond what is available on competing products. This allows customers to easily create new features similar to developing apps for mobile devices while taking full advantage of the system's performance, including full CPU and GPU power and programmable logic.