MementoGUI enhances long-horizon GUI agents with learned multimodal memory control, improving task success rates by over 20% on benchmarks.

Ziyun Zeng, Hang Hua, Bocheng Zou et al.

Decoupled semantic-spatial Transformer improves NVS fidelity by separating features, achieving 1.1dB PSNR gain with minimal latency.

Yihang Wu, Yihang Sun, Shaofeng Zhang et al.

StableHand estimates dual-hand motion from egocentric video using quality-aware flow matching, reducing W-MPJPE by 20-25%.

Huajian Zeng, Chaohua Yao, Yuantai Zhang et al.

SGSoft learns fused semantic-geometric features for 3D shape correspondence via template-guided soft signals, achieving state-of-the-art inter-category generalization.

Soyeon Yoon, Chang Wook Seo, Hyunjung Shim

Generation Navigator optimizes image generation with PRE-GRPO, achieving a WISE score of 0.90.

Jinming Liu, Ruoyu Feng, Yuqi Wang et al.

EPIC-Bench uses 6.6k annotated samples to systematically evaluate vision-language models' fine-grained embodied perception in real-world scenarios.

Haozhe Shan, Xiancong Ren, Han Dong et al.

VGGT-Occ enhances 3D occupancy prediction with geometric embedding and density-aware gated fusion, achieving 33.64% IoU.

Xun Chen, Tianchen Deng, Rui Wang et al.

WOW-Seg uses Mask2Token and Cascade Attention Mask to enable open-world segmentation with only 1/8 parameters of SOTA, achieving 89.7 semantic similarity.

Danyang Li, Tianhao Wu, Bin Li et al.

DriveSafe combines multimodal scene descriptions with LLMs for risk detection and safety suggestions, outperforming zero-shot models with 52.85% accuracy.

Sainithin Artham, Shankar Gangisetty, Avijit Dasgupta et al.

EgoKit offers unified low-cost egocentric data collection across six heterogeneous devices.

Liuchuan Yu, Erdem Murat, Beichen Wang et al.

EVA01 integrates 3D mesh understanding and generation using Mixture-of-Transformers.

Zongyuan Yang, Mingjing Yi, Wanli Ma et al.

VAGS adaptively adjusts guidance scale based on velocity direction similarity, improving image fidelity and semantic control without extra training.

Yan Luo, Ahmadou Aidara, Jingyi Lu et al.

DriveCtrl narrows the sim-to-real gap in driving videos using depth-conditioned generation.

Haonan Zhao, Yiting Wang, Jingkun Chen et al.

TAB-VLM tests 600 questions on 1,600 Indian artifacts; GPT-5.2 leads, yet reaches only 58.7% overall accuracy.

Mukul Ranjan, Prince Jha, Khushboo Kumari et al.

Proposed Nonlinear Bipolar Compensation method improves low-bit quantized model accuracy by 4.6% on average.

Peilin Sun, Jianxin Wu

Proposes a self-adaptive interleaved reasoning framework with hierarchical data pipeline, boosting multimodal generation fidelity and efficiency.

Qingyang Liu, Bingjie Gao, Canmiao Fu et al.

EponaV2 employs future depth and semantic prediction with flow matching to enhance autonomous driving planning, outperforming perception-free models with +1.3PDMS and +5.5EPDMS improvements.

Jiawei Xu, Zhizhou Zhong, Zhijian Shu et al.

TOPOS framework uses TOPOS-VAE and TOPOS-DiT for high-fidelity 3D head generation, outperforming existing methods.

Bojun Xiong, Zoubin Bi, Xinghui Peng et al.

CreFlow employs a compositional LTL reward model with localized correction and success-guided reflow loss, boosting sparse-reward embodied video RL success by 23.8%.

Zhenyang Ni, Yijiang Li, Ruochen Jiao et al.

Proposes SIAA gray-box attack exploiting ViT feature space, achieving near-white-box success rates (~98%) in fake image detection.

Chiara Musso, Joy Battocchio, Andrea Montibeller et al.