Large language models (LLMs) excel at generating empathic responses in text-based conversations. But, how reliably do they judge the nuances of empathic communication? We investigate this question by comparing how experts, crowdworkers, and LLMs annotate empathic communication across four evaluative frameworks drawn from psychology, natural language processing, and communications applied to 200 real-world conversations where one speaker shares a personal problem and the other offers support. Drawing on 3,150 expert annotations, 2,844 crowd annotations, and 3,150 LLM annotations, we assess inter-rater reliability between these three annotator groups. We find that expert agreement is high but varies across the frameworks' sub-components depending on their clarity, complexity, and subjectivity. We show that expert agreement offers a more informative benchmark for contextualizing LLM performance than standard classification metrics. Across all four frameworks, LLMs consistently approach this expert level benchmark and exceed the reliability of crowdworkers. These results demonstrate how LLMs, when validated on specific tasks with appropriate benchmarks, can support transparency and oversight in emotionally sensitive applications including their use as conversational companions.

The utilization of large-scale neural networks on Processing-In-Memory (PIM) accelerators encounters challenges due to constrained on-chip memory capacity. To tackle this issue, current works explore model compression algorithms to reduce the size of Convolutional Neural Networks (CNNs). Most of these algorithms either aim to represent neural operators with reduced-size parameters (e.g., quantization) or search for the best combinations of neural operators (e.g., neural architecture search). Designing neural operators to align with PIM accelerators' specifications is an area that warrants further study. In this paper, we introduce the Epitome, a lightweight neural operator offering convolution-like functionality, to craft memory-efficient CNN operators for PIM accelerators (EPIM). On the software side, we evaluate epitomes' latency and energy on PIM accelerators and introduce a PIM-aware layer-wise design method to enhance their hardware efficiency. We apply epitome-aware quantization to further reduce the size of epitomes. On the hardware side, we modify the datapath of current PIM accelerators to accommodate epitomes and implement a feature map reuse technique to reduce computation cost. Experimental results reveal that our 3-bit quantized EPIM-ResNet50 attains 71.59% top-1 accuracy on ImageNet, reducing crossbar areas by 30.65 times. EPIM surpasses the state-of-the-art pruning methods on PIM.

Peter Watts is the author of some of the darkest and most thoroughly researched science fiction novels ever written. One of his early fans was horror author Theresa DeLucci, who read his debut novel Starfish while working at Tor Books in the early 2000s. "I had never really read a lot of hard science fiction, but his concepts really intrigued me, and the editor at the time told me that it was really, really dark, and he thought that I would like it, and he was absolutely correct," DeLucci says in Episode 551 of the Geek's Guide to the Galaxy podcast. Watts is best known for his 2006 novel Blindsight, about a crew of augmented humans who are sent to intercept an alien vessel. Science fiction author Sam J. Miller says that Blindsight features some of the best-written aliens in all of science fiction.

In order to study the properties of total visual input in humans, a single subject wore a camera for two weeks capturing, on average, an image every 20 seconds (www.research.microsoft.com/ The resulting new dataset contains a mix of indoor and outdoor scenes as well as numerous foreground objects. Our first analysis goal is to create a visual summary of the subject's two weeks of life using unsupervised algorithms that would automatically discover recurrent scenes, familiar faces or common actions. Photosynth) or appearance-based clustering models (e.g. the epitome), is impractical due to either the large dataset size or the dramatic variation in the lighting conditions. As a remedy to these problems, we introduce a novel image representation, the "stel epitome," and an associated efficient learning algorithm.

We study the problem of segmenting specific white matter structures of interest from Diffusion Tensor (DT-MR) images of the human brain. This is an important requirement in many Neuroimaging studies: for instance, to evaluate whether a brain structure exhibits group level differences as a function of disease in a set of images. Typically, interactive expert guided segmentation has been the method of choice for such applications, but this is tedious for large datasets common today. To address this problem, we endow an image segmentation algorithm with'advice' encoding some global characteristics of the region(s) we want to extract. This is accomplished by constructing (using expert-segmented images) an epitome of a specific region - as a histogram over a bag of'words' (e.g.,suitable feature descriptors).

Variational autoencoder (VAE) is a widely used generative model for unsupervised learning of vector data. The learning capacity of VAE is often limited by \textit{over-pruning} - a phenomenon that prevents many of the latent dimensions from learning any useful information about the input data. Variational graph autoencoder (VGAE) extends VAE for unsupervised learning of graph-structured data. Being an extension of VAE model, VGAE, also suffers from over-pruning in principal. In this paper we look at over-pruning in VGAE and observe that the generative capacity of VGAE is limited because of the way VGAE deals with this issue. We then propose epitomic variational graph autoencoder (EVGAE), a generative variational framework for graph datasets to overcome over-pruning. We show through experiments that the resulting model has a better generative ability and also achieves better scores in graph analysis related tasks.

We study the problem of segmenting specific white matter structures of interest from Diffusion Tensor (DT-MR) images of the human brain. This is an important requirement in many Neuroimaging studies: for instance, to evaluate whether a brain structure exhibits group level differences as a function of disease in a set of images. Typically, interactive expert guided segmentation has been the method of choice for such applications, but this is tedious for large datasets common today. To address this problem, we endow an image segmentation algorithm with'advice' encoding some global characteristics of the region(s) we want to extract. This is accomplished by constructing (using expert-segmented images) an epitome of a specific region - as a histogram over a bag of'words' (e.g.,suitable feature descriptors).

In order to study the properties of total visual input in humans, a single subject wore a camera for two weeks capturing, on average, an image every 20 seconds (www.research.microsoft.com/ The resulting new dataset contains a mix of indoor and outdoor scenes as well as numerous foreground objects. Our first analysis goal is to create a visual summary of the subject's two weeks of life using unsupervised algorithms that would automatically discover recurrent scenes, familiar faces or common actions. Photosynth) or appearance-based clustering models (e.g. the epitome), is impractical due to either the large dataset size or the dramatic variation in the lighting conditions. As a remedy to these problems, we introduce a novel image representation, the "stel epitome," and an associated efficient learning algorithm.

We said last year that Sony had put Bose "on notice" when it comes to active noise-cancelling headphones. Our review of Sony's WH-1000XM2 reported that Sony not only delivered incredible audio quality, but that the company offered some high-tech features Bose couldn't match. These headphones are superior to the Bose QuietComfort 35 II in almost every way. Sony retained all the features that we liked in the previous iteration, including adaptive sound control, gesture recognition, and great audio reproduction (at least when powered), and made significant improvements to its active noise-cancellation technology. Sony's new headphones are also more comfortable to wear for long listening sessions.

At SentinelOne, we often tout machine learning and behavioral detection as the epitome of malware prevention, mitigation, and remediation. This is true--but only because we do it the right way. Our systems are trained correctly and support by a host of interlocking features such as cloud intelligence. This is the second of a two-part series about the pros and cons of AI in security--and what might happen if it goes wrong. In our last article in this series, we looked a couple of ways in which hackers have tried to fool AI.