Abstract--Real-world machine learning applications are often hindered by two critical challenges: distribution shift and label noise. Networks inherently tend to overfit to redundant, uninformative features present in the training distribution, which undermines their ability to generalize effectively to the target domain's distribution. The presence of noisy data further exacerbates this issue by inducing additional overfitting to noise, causing existing domain generalization methods to fail in effectively distinguishing invariant features from spurious ones. We also introduce a weighted loss function that dynamically adjusts the contribution of each sample based on its distance to the corresponding NLP anchor, thereby improving the model's resilience to noisy labels. Generalization (DG) has emerged as a pivotal algorithm in machine learning, aiming to develop models that can maintain high performance on previously unseen environments--or domains. T raditional methods often assume that training and test data share the same distribution, yet in real-world scenarios, there is frequently a substantial shift between these distributions. This phenomenon, widely referred to as domain shift, can cause severe performance degradation in tasks spanning computer vision, natural language processing, and medical image analysis [1]. As shown in Figure 1(a)(b), even within the same class label, the distribution of feature representations can vary considerably . This variation may stem from differences in image acquisition conditions--such as lighting variations, changes in pose, or complex background environments--and even from more subtle domain-specific factors like sensor noise or camera calibration differences. Such intra-class variability poses a significant challenge for developing accurate and adaptable models, which must learn to extract invariant features that capture the true semantic essence of the class while ignoring irrelevant variations. Lin, Z. Zhang is with Worcester Polytechnic Institute, Worcester, MA, 01890. L.Wang is with Carnegie Mellon University, Pittsburgh, P A, 15213. Y .Wang is with Peking University, Beijing, China, 100871. Z.Li, W.Lu is with T singhua University, Beijing, China, 100190. K.Y amada is with T ohoku University, Sendai, Japan, 980-8572.

Navigating the ultrasound (US) probe to the standardized imaging plane (SIP) for image acquisition is a critical but operator-dependent task in conventional freehand diagnostic US. Robotic US systems (RUSS) offer the potential to enhance imaging consistency by leveraging real-time US image feedback to optimize the probe pose, thereby reducing reliance on operator expertise. However, determining the proper approach to extracting generalizable features from the US images for probe pose adjustment remain challenging. In this work, we propose a SIP navigation framework for RUSS, exemplified in the context of robotic lung ultrasound (LUS). This framework facilitates automatic probe adjustment when in proximity to the SIP. This is achieved by explicitly extracting multiple anatomical features presented in real-time LUS images and performing non-patient-specific template matching to generate probe motion towards the SIP using image-based visual servoing (IBVS). This framework is further integrated with the active-sensing end-effector (A-SEE), a customized robot end-effector that leverages patient external body geometry to maintain optimal probe alignment with the contact surface, thus preserving US signal quality throughout the navigation. The proposed approach ensures procedural interpretability and inter-patient adaptability. Validation is conducted through anatomy-mimicking phantom and in-vivo evaluations involving five human subjects. The results show the framework's high navigation precision with the probe correctly located at the SIP for all cases, exhibiting positioning error of under 2 mm in translation and under 2 degree in rotation. These results demonstrate the navigation process's capability to accomondate anatomical variations among patients.

A game's theme is an important part of its design -- it conveys narrative information, rhetorical messages, helps the player intuit strategies, aids in tutorialisation and more. Thematic elements of games are notoriously difficult for AI systems to understand and manipulate, however, and often rely on large amounts of hand-written interpretations and knowledge. In this paper we present a technique which connects game embeddings, a recent method for modelling game dynamics from log data, and word embeddings, which models semantic information about language. We explain two different approaches for using game embeddings in this way, and show evidence that game embeddings enhance the linguistic translations of game concepts from one theme to another, opening up exciting new possibilities for reasoning about the thematic elements of games in the future.

Large Language Models (LLMs) are notorious for blending fact with fiction and generating non-factual content, known as hallucinations. To tackle this challenge, we propose an interactive system that helps users obtain insights into the reliability of the generated text. Our approach is based on the idea that the self-consistency of multiple samples generated by the same LLM relates to its confidence in individual claims in the generated texts. Using this idea, we design RELIC, an interactive system that enables users to investigate and verify semantic-level variations in multiple long-form responses. This allows users to recognize potentially inaccurate information in the generated text and make necessary corrections. From a user study with ten participants, we demonstrate that our approach helps users better verify the reliability of the generated text. We further summarize the design implications and lessons learned from this research for inspiring future studies on reliable human-LLM interactions.

Anti-surveillance makeup, used by people who do not want to be identified to fool facial recognition systems, is bold and striking, not exactly the stuff of cloak and daggers. While experts' opinions vary on the makeup's effectiveness to avoid detection, they agree that its use is not yet widespread. Anti-surveillance makeup relies heavily on machine learning and deep learning models to "break up the symmetry of a typical human face" with highly contrasted markings, says John Magee, an associate computer science professor at Clark University in Worcester, MA, who specializes in computer vision research. However, Magee adds that "If you go out [wearing] that makeup, you're going to draw attention to yourself." The effectiveness of anti-surveillance makeup has been debated because of racial justice protesters who do not want to be tracked, Magee notes.

Chowbotics is packing up Sally the salad making robot and sending it off to college. Well, many colleges actually, as the food robotics startup is set to announce next week a bigger push into the higher education market. Chowbotics told us that this school year, students at multiple colleges and universities in the U.S. will be able to buy salads and breakfast bowls from Sally the robot. Those schools include: Case Western Reserve University in Cleveland, OH; College of the Holy Cross in Worcester, MA; the University of Guelph in Ontario, Canada; Elmira College in Elmira, NY; the University of Memphis in Memphis, TN; and Wichita State University in Wichita, KS. These schools join Marshall University in Huntington, WV, which installed Sally in 2018.

Jon Oltsik, an analyst at Enterprise Strategy Group in Milford, Mass., looked into potential pitfalls associated with enterprise security teams collecting ever-increasing reams of data. ESG research indicates that 38% of organizations collect more than 10 TB of data every month, primarily from firewall logs, network devices, antivirus and user activity logs. "Let's face it, well-intentioned security teams are being buried by data today. They go through heroic efforts and do what they can, but there is an obvious and logical outcome here: As security data volume grows, security professionals will only be able to derive an incremental amount of value," Oltsik said.

Cybersecurity machine learning is growing in popularity, according to Jon Oltsik, an analyst with Enterprise Strategy Group Inc. in Milford, Mass. Oltsik attended the recent Black Hat conference, where technology vendors were abuzz with talk of cybersecurity machine learning. ESG research asked 412 respondents about their understanding of artificial intelligence (AI) and cybersecurity machine learning, which revealed that only 30% said they were very knowledgeable on the subject. Only 12% of respondents said their organizations had deployed these systems widely. According to Olstik, the cybersecurity industry sees an opportunity, because only 6% of respondents in surveys said their organizations were not considering AI or machine learning deployments.