Introduces a time-localized parametric decomposition method for respiratory airflow analysis, achieving reconstruction error <0.001.

Victoria Ribeiro Rodrigues, Paul W. Davenport, Nicholas J. Napoli

CRAFT method enhances BLEU score by 2.13 points in English-Hindi translation through clustered regression for adaptive filtering.

Parthasarathi Panda, Asheswari Swain, Subhrakanta Panda

SS3D achieves end-to-end self-supervised 3D estimation from monocular video using the YouTube-8M dataset.

Marwane Hariat, Gianni Franchi, David Filliat et al.

BERAG improves retrieval-augmented generation with Bayesian ensemble, significantly enhancing knowledge-based visual question answering performance.

Jinghong Chen, Jingbiao Mei, Guangyu Yang et al.

WG-SRC provides operational feature fingerprints for graph datasets using a white-box signal-subspace probe, enhancing node classification accuracy.

Yuchen Xiong, Swee Keong Yeap, Zhen Hong Ban

Evaluates eight Shapley variants' human utility in high-risk settings, revealing misalignment between current metrics and human perception.

Inês Oliveira e Silva, Sérgio Jesus, Iker Perez et al.

Research evaluates QPP for selecting the best query variant in RAG pipelines to enhance generation quality.

Negar Arabzadeh, Andrew Drozdov, Michael Bendersky et al.

PASR achieves 81.59% Top-1 retrieval accuracy on Pix3D and 76.43% on Pascal3D datasets.

Jiaxin Shi, Guofeng Zhang, Wufei Ma et al.

Utilizing TARA framework for non-invasive 3D identification of group-housed livestock, achieving 100% accuracy.

Shiva Paudel, TsungCheng Tsai, Dongyi Wang

The Structure-Guided Diffusion Model (SGDM) integrates structural information to enhance EEG-based visual reconstruction fidelity.

Yongxiang Lian, Yueyang Cang, Pingge Hu et al.

CLVAE model uses a variational autoencoder for long-term customer revenue forecasting, enhancing accuracy.

Jeffrey Näf, Riana Valera Mbelson, Markus Meierer

The Mixed Membership sub-Gaussian Model (MMSG) addresses the limitation of classical GMM by allowing observations to belong to multiple components.

Huan Qing

GazeVLA learns human intention to enhance robotic manipulation, significantly outperforming baseline methods.

Chengyang Li, Kaiyi Xiong, Yuan Xu et al.

From natural language to verified code using Dafny, Gemma 4-31B achieved a 90.91% verification success rate.

Md Erfan, Md Kamal Hossain Chowdhury, Ahmed Ryan et al.

Proposes an LLM-based evaluation framework to enhance math reasoning assessment accuracy beyond symbolic math limitations.

Erez Yosef, Oron Anschel, Shunit Haviv Hakimi et al.

EV-CLIP efficiently adapts CLIP for few-shot action recognition under visual challenges using visual prompts.

Hyo Jin Jon, Longbin Jin, Eun Yi Kim

RedVLA identifies physical safety risks in VLA models through a two-stage process, achieving an ASR of 95.5%.

Yuhao Zhang, Borong Zhang, Jiaming Fan et al.

FlowAnchor stabilizes video editing signals using spatial attention and adaptive modulation for efficient multi-object scene editing.

Ze Chen, Lan Chen, Yuanhang Li et al.

WassersteinGrad explains dynamic physical field predictions by computing the entropic Wasserstein barycenter, enhancing autoregressive weather forecasting model interpretability.

Younes Essafouri, Laure Raynaud, Luciano Drozda et al.

The study shows that deep networks learn useful nonrobust features in medical images, achieving high accuracy on five MedMNIST tasks.

Coenraad Mouton, Randle Rabe, Niklas C. Koser et al.