Introduced NEMO-DE and NEEF-DE evolutionary frameworks for near-field multi-source localization, avoiding grid mismatch errors.

Seyed Jalaleddin Mousavirad, Parisa Ramezani, Mattias O'Nils et al.

Proposes a write–manage–read-based multi-dimensional memory framework, significantly enhancing LLM agent long-term memory capabilities.

Pengfei Du

UniLongGen enhances long-sequence image generation stability via dynamic memory management.

Haoyu Chen, Qing Liu, Yuqian Zhou et al.

InterReal combines physics simulation and automatic reward learning to achieve high-precision human-object interaction control in humanoid robots, outperforming baselines.

Dayang Liang, Yuhang Lin, Xinzhe Liu et al.

UnSCAR employs a multi-expert MoE architecture, supporting over 16 degradation types, enhancing scalability and controllability.

Debabrata Mandal, Soumitri Chattopadhyay, Yujie Wang et al.

Improving reasoning performance via uncertainty minimization, excelling on MATH500 and GSM8K datasets.

Nicolas Legrand, Kenneth Enevoldsen, Márton Kardos et al.

LiveWorld addresses 'out-of-sight dynamics' by introducing a monitor-based mechanism for persistent world evolution.

Zicheng Duan, Jiatong Xia, Zeyu Zhang et al.

Entropy-Aware On-Policy Distillation balances mode-seeking and diversity, boosting math reasoning accuracy by +1.37 to +5.05 on benchmarks.

Woogyeol Jin, Taywon Min, Yongjin Yang et al.

Optimizes image-based shape retrieval using pre-aligned encoders and hard contrastive learning, achieving top performance with OpenShape and Point-BERT across datasets.

Paul Julius Kühn, Cedric Spengler, Michael Weinmann et al.

Swimba enhances SSM capacity by mixing expert streams in parameter space, maintaining computational efficiency.

Zhixu Du, Krishna Teja Chitty-Venkata, Murali Emani et al.

Proposes a design procedure-based prompting strategy using LLMs to generate industrial-grade CAD programs with high geometric diversity for data augmentation.

Yan-Ying Chen, Dule Shu, Matthew Hong et al.

Proposes a passive background subtraction method to estimate each diffuse LiDAR pixel’s footprint and spatial sensitivity in RGB images.

Nikhil Behari, Ramesh Raskar

Proposes WanderDream, a world model-based emulative framework generating 15.8K videos and 158K QA pairs for scene reasoning without active exploration.

Ruiping Liu, Yufan Chen, Yuheng Zhang et al.

LATO uses topology-preserving sparse voxel VAE to efficiently generate structured 3D meshes.

Tianhao Zhao, Youjia Zhang, Hang Long et al.

Proposes auto aggregation and per-pixel rescaling to enhance training-free diffusion segmentors, leveraging stronger generative models for improved semantic segmentation.

Benyuan Meng, Qianqian Xu, Zitai Wang et al.

Proposes Skeleton-to-Image encoding (S2I) enabling self-supervised skeleton representation learning via vision-pretrained models, improving cross-format generalization.

Siyuan Yang, Jun Liu, Hao Cheng et al.

LongNAP combines parametric and retrieval mechanisms, trained with policy gradients on 360K actions over 1,800 hours, outperforming baselines significantly.

Omar Shaikh, Valentin Teutschbein, Kanishk Gandhi et al.

InnoAds-Composer employs single-stage tri-conditional diffusion with importance routing, achieving 54.39 FID and 0.857 sentence accuracy, greatly improving e-commerce poster generation.

Yuxin Qin, Ke Cao, Haowei Liu et al.

StreamWise employs adaptive scheduling on heterogeneous hardware to generate 10-minute podcast videos in 1.4 hours at under $25, with response times in seconds.

Haoran Qiu, Gohar Irfan Chaudhry, Chaojie Zhang et al.

Contact-Grounded Policy achieves superior dexterous manipulation via generative contact prediction, outperforming existing baselines.

Zhengtong Xu, Yeping Wang, Ben Abbatematteo et al.