This paper introduces BIM-informed visual SLAM, reducing trajectory error by 25.23% and map inaccuracies by 7.14%, leveraging structural priors for construction site mapping.

Asier Bikandi-Noya, Miguel Fernandez-Cortizas, Muhammad Shaheer et al.

Proposes Any2Track, a two-stage RL framework enabling humanoid robots to robustly track diverse motions under multiple disturbances with zero-shot sim2real transfer.

Zhikai Zhang, Jun Guo, Chao Chen et al.

Proposes geometric volume and suspicion metrics for improved hallucination detection and correction in LLMs.

Edward Phillips, Sean Wu, Soheila Molaei et al.

UM-Depth enhances monocular depth estimation with motion and uncertainty, excelling on KITTI datasets.

Tae-Wook Um, Ki-Hyeon Kim, Hyun-Duck Choi et al.

GenPAS framework improves generative recommendation accuracy via three-step sampling, showing up to 783.7% performance boost.

Geon Lee, Bhuvesh Kumar, Clark Mingxuan Ju et al.

WebWeaver employs dynamic outlines and dual agents, significantly improving open-ended deep research accuracy and efficiency.

Zijian Li, Xin Guan, Bo Zhang et al.

Proposes Agentic CPT framework, leveraging large-scale multi-task data to enhance foundational models' agentic abilities, achieving 39.9% on BrowseComp-en.

Liangcai Su, Zhen Zhang, Guangyu Li et al.

WebResearcher employs iterative research and WebFrontier data engine to achieve state-of-the-art long-horizon reasoning, surpassing 6 benchmarks with significant improvements.

Zile Qiao, Guoxin Chen, Xuanzhong Chen et al.

Proposes a deep learning-based co-pilot framework enabling real-time generation of task-specific robot morphologies, reducing design time from hours to seconds.

Jonathan Külz, Sehoon Ha, Matthias Althoff

Proposes a Graph-Latent Neural ODE framework combining GNNs, NODEs, and neural processes for cardiac trajectory reconstruction and classification.

Jaume Banus, Augustin C. Ogier, Roger Hullin et al.

Conan-Embedding-v2 is a 1.4B parameter from-scratch trained model, using soft masks and dynamic hard negative mining to achieve SOTA text embeddings.

Shiyu Li, Yang Tang, Ruijie Liu et al.

Using linear discriminant analysis (LDA), the study identifies orthogonal directions encoding tense and aspect in LLM residual space, enabling causal control during multi-token generation.

Alina Klerings, Jannik Brinkmann, Daniel Ruffinelli et al.

EgoMem enables personalized dialogues with full-duplex models, achieving over 95% accuracy in user recognition.

Yiqun Yao, Naitong Yu, Xiang Li et al.

AgenticIE system extracts information from complex regulatory documents, achieving a 16% EM score improvement.

Gaye Colakoglu, Gürkan Solmaz, Jonathan Fürst

This survey explores reasoning topologies in time series with large language models, introducing three structures.

Ching Chang, Yidan Shi, Defu Cao et al.

Proposes a graph-based System Error Propagation Graph (SEPG) to analyze end-to-end sensor attack impacts, revealing 12 overlooked attack vectors.

Qingzhao Zhang, Shaocheng Luo, Z. Morley Mao et al.

Proposes a kernel-based stochastic approximation framework for nonlinear operator learning in infinite-dimensional spaces, with dimension-free convergence guarantees.

Jia-Qi Yang, Lei Shi

ImMimic combines DTW mapping and MixUp, reaching 1.00 success on several tasks with only five robot demonstrations.

Yangcen Liu, Woo Chul Shin, Yunhai Han et al.

Maestro employs multi-agent self-evolution with Critic and Judge modules, leveraging pairwise preferences to enhance text-to-image generation quality autonomously.

Xingchen Wan, Han Zhou, Ruoxi Sun et al.

MagicMirror uses 340K annotated images and GRPO-trained VLM for fine-grained artifact assessment, advancing model defect detection.

Jia Wang, Jie Hu, Xiaoqi Ma et al.