Concatenating LLM-generated features to graph neural networks systematically reduces accuracy on homophilous benchmarks, with PubMed accuracy dropping by -17.0 ± 0.3 pp. This degradation is linked to LLM-alone discriminability (Delta_sig), which correlates strongly with concatenation cost (r² = 0.38) and shows a power law relationship with feature dimension and node count (r² = 0.97), particularly in low-Delta_sig, low-node scenarios.
Stanford introduces SEFD, an open, layout-faithful reconstruction of SEC filings into MultiMarkdown. The 152B-token SEFD-v1 dataset enables financial language modeling and includes benchmarks for forecasting and table transcription, with less than 0.1% overlap to Common Crawl.
FusionRS introduces the first large-scale RGB-infrared-text dataset for remote sensing vision-language modeling. It aligns RGB and infrared images with IR-aware captions, enabling dual-modal vision-language foundation models. Experiments show improved RGB-IR alignment, retrieval, and captioning, with ablation studies confirming the critical role of modality-specific textual supervision.
A study proposes a hybrid Ret-DNN with XGBoost model to forecast customer behavior in e-commerce. Using 500,000 transaction records from a UK retailer, the model achieves a Mean Absolute Error of 0.2193, outperforming the existing Ret-DNN model.
McWC introduces a model that separately captures cyclicity, trend, and inter-channel correlations in long-term time series forecasting. It uses multi-layer cyclicity construction, wavelet decomposition, and a multi-layer perceptron to extract and fuse high- and low-frequency information, while decoupling intra-channel autocorrelations via frequency-domain loss. Experiments on six real-world datasets show McWC achieves state-of-the-art performance with high computational efficiency.
McWC introduces a model that separately captures cyclicity, trend, and inter-channel correlations in long-term time series forecasting. It uses multi-layer cyclicity construction, wavelet decomposition, and a multi-layer perceptron to extract and fuse high- and low-frequency information, while decoupling intra-channel autocorrelations via frequency-domain loss. Experiments on six real-world datasets show McWC achieves state-of-the-art performance with high computational efficiency.
The Certus Caliber Classification Gunshot Dataset (C3GD) contains over 8000 field-collected gunshot audio samples from 28 firearms across 16 calibers. It offers detailed metadata on firearms, calibers, microphones, and placement, enabling robust academic analysis and real-world applications in gunshot detection and audio signal processing.
This study evaluates Word2Vec's ability to capture semantic relationships in Toki Pona, a language with only 130 words. Using 1.4 million sentences, it finds that non-core tokens do not disrupt embedding structure and may actually bring similar words closer in vector space. The results show Word2Vec's effectiveness relies more on distributional patterns than vocabulary size, even at extreme lexical reduction.
MultiClin introduces a clinical ASR benchmark that evaluates models' robustness to multiscript variability. It shows that multiscript-aware evaluation outperforms conventional single-reference methods, and script unification yields the best ASR performance, while inconsistent script mappings increase orthographic uncertainty.
IMPACTeen is a dataset of 1,021 texts annotated from five perspectives—teenagers, parents, psychologists, communication experts, and teachers. It includes 5,100 annotation records covering social influence techniques, intentions, consequences, and resistance, with annotations validated through human editing. The dataset, created using LLM generation and human validation, is available in both Polish and English and supports research on social influence and language model training.
IMPACTeen is a dataset of 1,021 texts annotated from five perspectives—teenagers, parents, psychologists, communication experts, and teachers. It includes 5,100 annotation records covering social influence techniques, intentions, consequences, and resistance, with annotations validated through human editing. The dataset, created using LLM generation and human validation, is available in both Polish and English and supports research on social influence and language model training.
A study finds that while textual review signals can be fused with collaborative data, their marginal contribution remains limited compared to collaborative signals in matrix factorization models. Adaptive fusion and cross-attention mechanisms improve representation flexibility, but do not significantly boost performance across datasets.
A new method decouples ML inference from state persistence in streaming systems using probabilistic thinning. It selectively triggers durable state updates based on event informativeness, reducing persistence path overhead by up to 90% without compromising downstream utility or introducing systemic errors.
This survey presents a mathematical framework for analyzing geometric data, integrating differential geometry, statistics, and machine learning. It outlines a unified pipeline for shape representation, geodesic metrics, statistical analysis, and geometry-aware learning, enabling the study of shape variability and structural trajectories across populations and time. Applications span biology, medicine, anthropology, and computer vision, highlighting challenges in handling nonlinear and unaligned geometric variation.