arXiv:2510.21742v1 Announce Type: new Abstract: The statistics of correlations are central quantities characterizing the collective dynamics of recurrent neural networks. We derive exact expressions for the statistics of correlations of nonlinear recurrent networks in the limit of a large number N of neurons, including systematic 1/N corrections. Our approach uses a path-integral representation of the […]
Reconnaissance Automatique des Langues des Signes : Une Approche Hybrid’ee CNN-LSTM Bas’ee sur Mediapipe
arXiv:2510.22011v1 Announce Type: cross Abstract: Sign languages play a crucial role in the communication of deaf communities, but they are often marginalized, limiting access to essential services such as healthcare and education. This study proposes an automatic sign language recognition system based on a hybrid CNN-LSTM architecture, using Mediapipe for gesture keypoint extraction. Developed with […]
Active Hydrodynamic Theory of Euchromatin and Heterochromatin
arXiv:2503.20964v3 Announce Type: replace-cross Abstract: The genome contains genetic information essential for cell’s life. The genome’s spatial organization inside the cell nucleus is critical for its proper function including gene regulation. The two major genomic compartments — euchromatin and heterochromatin — contain largely transcriptionally active and silenced genes, respectively, and exhibit distinct dynamics. In this […]
Discovering Latent Graphs with GFlowNets for Diverse Conditional Image Generation
arXiv:2510.22107v1 Announce Type: cross Abstract: Capturing diversity is crucial in conditional and prompt-based image generation, particularly when conditions contain uncertainty that can lead to multiple plausible outputs. To generate diverse images reflecting this diversity, traditional methods often modify random seeds, making it difficult to discern meaningful differences between samples, or diversify the input prompt, which […]
Performance Trade-offs of Optimizing Small Language Models for E-Commerce
arXiv:2510.21970v1 Announce Type: new Abstract: Large Language Models (LLMs) offer state-of-the-art performance in natural language understanding and generation tasks. However, the deployment of leading commercial models for specialized tasks, such as e-commerce, is often hindered by high computational costs, latency, and operational expenses. This paper investigates the viability of smaller, open-weight models as a resource-efficient […]
Estimating the Error of Large Language Models at Pairwise Text Comparison
arXiv:2510.22219v1 Announce Type: cross Abstract: We measure LLMs’ output error at pairwise text comparison, noting the probability of error in their preferences. Our method does not rely on the ground truth and supports two scenarios: (i) uniform error rate regardless of the order of comparison, estimated with two comparisons for each text pair with either […]
Preference Optimization by Estimating the Ratio of the Data Distribution
arXiv:2505.19601v2 Announce Type: replace-cross Abstract: Direct preference optimization (DPO) is widely used as a simple and stable method for aligning large language models (LLMs) with human preferences. This paper investigates a generalized DPO loss that enables a policy model to match the target policy from a likelihood ratio estimation perspective. The ratio of the target […]
Predicting Metabolic Dysfunction-Associated Steatotic Liver Disease using Machine Learning Methods
arXiv:2510.22293v1 Announce Type: cross Abstract: Background: Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) affects ~33% of U.S. adults and is the most common chronic liver disease. Although often asymptomatic, progression can lead to cirrhosis. Early detection is important, as lifestyle interventions can prevent disease progression. We developed a fair, rigorous, and reproducible MASLD prediction model and […]
Distribution Shift Alignment Helps LLMs Simulate Survey Response Distributions
arXiv:2510.21977v1 Announce Type: new Abstract: Large language models (LLMs) offer a promising way to simulate human survey responses, potentially reducing the cost of large-scale data collection. However, existing zero-shot methods suffer from prompt sensitivity and low accuracy, while conventional fine-tuning approaches mostly fit the training set distributions and struggle to produce results more accurate than […]
Top-Down Semantic Refinement for Image Captioning
arXiv:2510.22391v1 Announce Type: cross Abstract: Large Vision-Language Models (VLMs) face an inherent contradiction in image captioning: their powerful single-step generation capabilities often lead to a myopic decision-making process. This makes it difficult to maintain global narrative coherence while capturing rich details, a limitation that is particularly pronounced in tasks that require multi-step and complex scene […]