AutoPP integrates unified design, element rendering, and IDPO-based CTR optimization, achieving state-of-the-art automatic product poster generation and online CTR improvement.

Jiahao Fan, Yuxin Qin, Wei Feng et al.

Co-GRPO optimizes both model and inference schedule, enhancing MDM generation quality.

Renping Zhou, Zanlin Ni, Tianyi Chen et al.

GeCo evaluates geometric consistency in video generation by fusing motion and depth priors to reveal distortions and occlusion inconsistencies.

Leslie Gu, Junhwa Hur, Charles Herrmann et al.

Utilized Transformer-based models and transfer learning to enhance NLP for six Senegalese languages, achieving up to 78% accuracy in text classification and BLEU scores of 32 in translation.

Derguene Mbaye, Tatiana D. P. Mbengue, Madoune R. Seye et al.

kooplearn is a Scikit-Learn compatible library for evolution operator learning, supporting linear, kernel, and deep learning estimators.

Giacomo Turri, Grégoire Pacreau, Giacomo Meanti et al.

LookPlanGraph uses VLM graph augmentation for instruction following in dynamic environments, improving task success rates.

Anatoly O. Onishchenko, Alexey K. Kovalev, Aleksandr I. Panov

Axlerod integrates LLM and RAG to achieve 93.18% accuracy in insurance policy retrieval, reducing search time by 2.42 seconds.

Adam Bradley, John Hastings, Khandaker Mamun Ahmed

A new ALE-based variable importance method addressing correlation bias, achieving higher reliability and computational efficiency.

Jingyu Zhu, Daniel W. Apley

Survey of 242 studies on LLM-based agentic software issue resolution, highlighting reinforcement learning's role and performance metrics.

Zhonghao Jiang, David Lo, Zhongxin Liu

GateBreaker is a training-free, lightweight attack framework targeting MoE LLMs, increasing attack success rate from 7.4% to 64.9%.

Lichao Wu, Sasha Behrouzi, Mohamadreza Rostami et al.

AutoHet system automates 3D parallelism on heterogeneous GPUs, optimizing device grouping and workload balancing, achieving up to 1.79× training speedup.

Yuxiao Wang, Yuedong Xu, Qingyang Duan et al.

TS-Arena employs a live forecasting protocol with pre-registration to prevent data leakage, enabling real-time evaluation of TSFMs with dynamic rankings.

Marcel Meyer, Sascha Kaltenpoth, Henrik Albers et al.

SemanticGen uses a two-stage diffusion framework in semantic space, boosting long video generation speed and quality, with a 30% faster convergence and better long-term consistency.

Jianhong Bai, Xiaoshi Wu, Xintao Wang et al.

This study systematically analyzes video diffusion Transformer (DiT) features for robust point tracking, outperforming ResNet, and introduces DiTracker for efficient zero-shot tracking.

Soowon Son, Honggyu An, Jisu Nam et al.

Fusion of YOLO-based gate detection and Kalman filter drift correction enhances monocular VIO for drone racing, achieving speeds over 59 km/h.

Maulana Bisyir Azhari, Donghun Han, Je In You et al.

Proposes QCU-Net, a hybrid quantum-classical diffusion model for high-quality labeled remote sensing image synthesis, achieving 64% FID reduction and enhanced semantic accuracy.

Francesco Mauro, Francesca De Falco, Lorenzo Papa et al.

CRAFT framework improves multimodal text-to-image generation via continuous reasoning and feedback tuning for enhanced accuracy and control.

V. Kovalev, A. Kuvshinov, A. Buzovkin et al.

ActionFlow system accelerates edge VLA inference by 2.55× using cross-request pipelining, no retraining needed.

Yuntao Dai, Hang Gu, Teng Wang et al.

DDAVS achieves state-of-the-art audio-visual segmentation using disentangled audio semantics and delayed bidirectional alignment on AVS-Objects and VPO benchmarks.

Jingqi Tian, Yiheng Du, Haoji Zhang et al.

Memory-T1 employs reinforcement learning for time-aware memory retrieval, boosting multi-session dialogue reasoning to 67.0% accuracy, outperforming baselines by over 10%.

Yiming Du, Baojun Wang, Yifan Xiang et al.