Memorization in language models is a critical yet poorly understood phenomenon. In this work, we investigate memorization in transformer-based language models by analyzing their memorization dynamics during training over multiple epochs. We find that memorization is neither a constant accumulation of sequences nor simply dictated by the recency of exposure to these sequences. Instead, much like generalization, memorization appears to be driven by pattern recognition. Tracking memorization dynamics in mixed datasets, we observe that models memorize different sub-datasets in distinct bursts, suggesting that each subset is associated with unique underlying patterns, and that the model prefers to learn these patterns in a consistent order. We also find that easily learnable patterns tend to support generalization on unseen data, while more complex patterns do not. Furthermore, in datasets with weak or absent patterns, larger models may delay memorization relative to smaller ones, a behavior we term overthinking. Our results show that the subset of sequences memorized by a model over time is not arbitrary, and give insights into the internal processes a model goes through during training.

Assessing software test coverage at scale remains a bottleneck in QA pipelines. We present LLM-as-a-Judge (LAJ), a production-ready, rubric-driven framework for evaluating Gherkin acceptance tests with structured JSON outputs. Across 20 model configurations (GPT-4, GPT-5 with varying reasoning effort, and open-weight models) on 100 expert-annotated scripts over 5 runs (500 evaluations), we provide the first comprehensive analysis spanning accuracy, operational reliability, and cost. We introduce the Evaluation Completion Rate (ECR@1) to quantify first-attempt success, revealing reliability from 85.4% to 100.0% with material cost implications via retries. Results show that smaller models can outperform larger ones: GPT-4o Mini attains the best accuracy (6.07 MAAE), high reliability (96.6% ECR@1), and low cost ($1.01 per 1K), yielding a 78x cost reduction vs. GPT-5 (high reasoning) while improving accuracy. Reasoning effort is model-family dependent: GPT-5 benefits from increased reasoning (with predictable accuracy-cost tradeoffs), whereas open-weight models degrade across all dimensions as reasoning increases. Overall, cost spans 175x ($0.45-$78.96 per 1K). We release the dataset, framework, and code to support reproducibility and deployment.

Large Language Model (LLM)-based conversational agents offer promising solutions for mental health support, but lack cultural responsiveness for diverse populations. This study evaluated the effectiveness of cultural prompting in improving cultural responsiveness and perceived empathy of LLM-generated therapeutic responses for Chinese American family caregivers. Using a randomized controlled experiment, we compared GPT-4o and Deepseek-V3 responses with and without cultural prompting. Thirty-six participants evaluated input-response pairs on cultural responsiveness (competence and relevance) and perceived empathy. Results showed that cultural prompting significantly enhanced GPT-4o's performance across all dimensions, with GPT-4o with cultural prompting being the most preferred, while improvements in DeepSeek-V3 responses were not significant. Mediation analysis revealed that cultural prompting improved empathy through improving cultural responsiveness. This study demonstrated that prompt-based techniques can effectively enhance the cultural responsiveness of LLM-generated therapeutic responses, highlighting the importance of cultural responsiveness in delivering empathetic AI-based therapeutic interventions to culturally and linguistically diverse populations.

Real-time multimodal inference on resource-constrained edge devices is essential for applications such as autonomous driving, human-computer interaction, and mobile health. However, prior work often overlooks the tight coupling between sensing dynamics and model execution, as well as the complex inter-modality dependencies. In this paper, we propose MMEdge, an new on-device multi-modal inference framework based on pipelined sensing and encoding. Instead of waiting for complete sensor inputs, MMEdge decomposes the entire inference process into a sequence of fine-grained sensing and encoding units, allowing computation to proceed incrementally as data arrive. MMEdge also introduces a lightweight but effective temporal aggregation module that captures rich temporal dynamics across different pipelined units to maintain accuracy performance. Such pipelined design also opens up opportunities for fine-grained cross-modal optimization and early decision-making during inference. To further enhance system performance under resource variability and input data complexity, MMEdge incorporates an adaptive multimodal configuration optimizer that dynamically selects optimal sensing and model configurations for each modality under latency constraints, and a cross-modal speculative skipping mechanism that bypasses future units of slower modalities when early predictions reach sufficient confidence. We evaluate MMEdge using two public multimodal datasets and deploy it on a real-world unmanned aerial vehicle (UAV)-based multimodal testbed. The results show that MMEdge significantly reduces end-to-end latency while maintaining high task accuracy across various system and data dynamics.

Large language models (LLMs) are increasingly integrated into academic workflows, with many conferences and journals permitting their use for tasks such as language refinement and literature summarization. However, their use in peer review remains prohibited due to concerns around confidentiality breaches, hallucinated content, and inconsistent evaluations. As LLM-generated text becomes more indistinguishable from human writing, there is a growing need for reliable attribution mechanisms to preserve the integrity of the review process. In this work, we evaluate topic-based watermarking (TBW), a semantic-aware technique designed to embed detectable signals into LLM-generated text. We conduct a systematic assessment across multiple LLM configurations, including base, few-shot, and fine-tuned variants, using authentic peer review data from academic conferences. Our results show that TBW maintains review quality relative to non-watermarked outputs, while demonstrating robust detection performance under paraphrasing. These findings highlight the viability of TBW as a minimally intrusive and practical solution for LLM attribution in peer review settings.

The Cellular-Potts model is a powerful and ubiquitous framework for developing computational models for simulating complex multicellular biological systems. Cellular-Potts models (CPMs) are often computationally expensive due to the explicit modeling of interactions among large numbers of individual model agents and diffusive fields described by partial differential equations (PDEs). In this work, we develop a convolutional neural network (CNN) surrogate model using a U-Net architecture that accounts for periodic boundary conditions. We use this model to accelerate the evaluation of a mechanistic CPM previously used to investigate in vitro vasculogenesis. The surrogate model was trained to predict 100 computational steps ahead (Monte-Carlo steps, MCS), accelerating simulation evaluations by a factor of 590 times compared to CPM code execution. Over multiple recursive evaluations, our model effectively captures the emergent behaviors demonstrated by the original Cellular-Potts model of such as vessel sprouting, extension and anastomosis, and contraction of vascular lacunae. This approach demonstrates the potential for deep learning to serve as efficient surrogate models for CPM simulations, enabling faster evaluation of computationally expensive CPM of biological processes at greater spatial and temporal scales.

Abstract--Running offers substantial health benefits, but improper gait patterns can lead to injuries, particularly without expert feedback. While prior gait analysis systems based on cameras, insoles, or body-mounted sensors have demonstrated effectiveness, they are often bulky and limited to offline, post-run analysis. Wrist-worn wearables offer a more practical and non-intrusive alternative, yet enabling real-time gait recognition on such devices remains challenging due to noisy Inertial Measurement Unit (IMU) signals, limited computing resources, and dependence on cloud connectivity. This paper introduces StrikeW atch, a compact wrist-worn system that performs entirely on-device, real-time gait recognition using IMU signals. As a case study, we target the detection of heel versus forefoot strikes to enable runners to self-correct harmful gait patterns through visual and auditory feedback during running. We propose four compact DL architectures (1D-CNN, 1D-SepCNN, LSTM, and Transformer) and optimize them for energy-efficient inference on two representative embedded Field-Programmable Gate Arrays (FPGAs): the AMD Spartan-7 XC7S15 and the Lattice iCE40UP5K. Using our custom-built hardware prototype, we collect a labeled dataset from outdoor running sessions and evaluate all models via a fully automated deployment pipeline. Our results reveal clear trade-offs between model complexity and hardware efficiency. Evaluated across 12 participants, 6-bit quantized 1D-SepCNN achieves the highest average F1 score of 0.847 while consuming just 0.350 µJ per inference with a latency of 0.140 ms on the iCE40UP5K running at 20 MHz. This configuration supports up to 13.6 days of continuous inference on a 320 mAh battery. Running is one of the most widely practiced sports worldwide, offering significant physical and mental benefits [1].

These authors contributed equally to this work. Abstract--The growing demand for smart home interfaces has increased interest in non-intrusive sensing methods like vibration-based gesture recognition. While prior studies demonstrated feasibility, they often rely on complex preprocessing and large Neural Networks (NNs) requiring costly high-performance hardware, resulting in high energy usage and limited real-world deployability. This study proposes an energy-efficient solution deploying compact NNs on low-power Field-Programmable Gate Arrays (FPGAs) to enable real-time gesture recognition with competitive accuracy. We adopt a series of optimizations: (1) We replace complex spectral preprocessing with raw waveform input, eliminating complex on-board preprocessing while reducing input size by 21 without sacrificing accuracy. A ping-pong buffering mechanism in 1D-SepCNN further improves deployability under tight memory constraints. Evaluated on two swipe-direction datasets with multiple users and ordinary tables, our approach achieves low-latency, energy-efficient inference on the AMD Spartan-7 XC7S25 FPGA. Under the PS data splitting setting, the selected 6-bit 1D-CNN reaches 0.970 average accuracy across users with 9.22 ms latency. The chosen 8-bit 1D-SepCNN further reduces latency to 6.83 ms (over 53 CPU speedup) with slightly lower accuracy (0.949). Both consume under 1.2 mJ per inference, demonstrating suitability for long-term edge operation.

The rapid growth of e-commerce has intensified the demand for Virtual Try-On (VTO) technologies, enabling customers to realistically visualize products overlaid on their own images. Despite recent advances, existing VTO models face challenges with fine-grained detail preservation, robustness to real-world imagery, efficient sampling, image editing capabilities, and generalization across diverse product categories. In this paper, we present DiT-VTON, a novel VTO framework that leverages a Diffusion Transformer (DiT), renowned for its performance on text-conditioned image generation, adapted here for the image-conditioned VTO task. We systematically explore multiple DiT configurations, including in-context token concatenation, channel concatenation, and ControlNet integration, to determine the best setup for VTO image conditioning. To enhance robustness, we train the model on an expanded dataset encompassing varied backgrounds, unstructured references, and non-garment categories, demonstrating the benefits of data scaling for VTO adaptability. DiT-VTON also redefines the VTO task beyond garment try-on, offering a versatile Virtual Try-All (VTA) solution capable of handling a wide range of product categories and supporting advanced image editing functionalities such as pose preservation, localized editing, texture transfer, and object-level customization. Experimental results show that our model surpasses state-of-the-art methods on VITON-HD, achieving superior detail preservation and robustness without reliance on additional condition encoders. It also outperforms models with VTA and image editing capabilities on a diverse dataset spanning thousands of product categories.