Mesh4D employs a VAE and diffusion model to reconstruct dynamic 3D meshes from monocular videos, capturing full shape and motion in a single pass.

Zeren Jiang, Chuanxia Zheng, Iro Laina et al.

DrivoR uses pretrained ViT and register tokens to compress multi-camera features, enabling efficient end-to-end autonomous driving with superior performance.

Ellington Kirby, Alexandre Boulch, Yihong Xu et al.

AlignXplore+ uses text as a universal interface for cross-task personalization, outperforming larger models on nine benchmarks with an 8B model.

Yuting Liu, Jian Guan, Jia-Nan Li et al.

SmartSearch enhances search agent query quality via process rewards and query refinement, significantly boosting retrieval efficiency.

Tongyu Wen, Guanting Dong, Zhicheng Dou

Proposes StealthGraph, using knowledge graphs to generate implicit harmful prompts; employs dual rewriting strategies to enhance stealth, achieving 84.92% attack success.

Huawei Zheng, Xinqi Jiang, Sen Yang et al.

HuForDet employs dual-branch architecture combining RGB/frequency face analysis and multimodal semantic consistency, achieving SOTA detection with 90.22% AUC.

Xiao Guo, Jie Zhu, Anil Jain et al.

RLVR training exhibits strong linear trajectories in parameters and outputs, enabling model extrapolation and 6.1× speedup via low-variance noise filtering.

Tianle Wang, Jiayu Liu, Zhongyuan Wu et al.

XGrammar-2 introduces TagDispatch and cross-grammar cache, achieving 6× faster structured generation for dynamic agent workloads.

Linzhang Li, Yixin Dong, Guanjie Wang et al.

Proposed a discrete Benamou-Brenier formulation on graphs to classify W1 geodesics.

Kieran Morris, Oliver Johnson

LocalDPO enhances video diffusion models by optimizing localized details, outperforming on Wan2.1 and CogVideoX.

Zitong Huang, Kaidong Zhang, Yukang Ding et al.

HoneyTrap employs multi-agent collaboration with delay, misdirection, and resource drain, reducing attack success rate by 68.77%.

Siyuan Li, Xi Lin, Jun Wu et al.

ResTok introduces hierarchical residual visual tokenizer, achieving gFID 2.34 with only 9 steps on ImageNet-256.

Xu Zhang, Cheng Da, Huan Yang et al.

Proposes TabTrim, transforming table pruning from sequential revisions to gold trajectory-supervised parallel search, boosting TableQA accuracy by 3.2%.

Yu Guo, Shenghao Ye, Shuangwu Chen et al.

This study evaluates whether LLM self-explanations improve users' ability to predict model behavior via counterfactual perturbations, showing consistent gains especially with pragmatic strategies.

Pingjun Hong, Benjamin Roth

O-Researcher uses multi-agent distillation and RL to generate high-quality research data, boosting open-source LLMs' performance.

Yi Yao, He Zhu, Piaohong Wang et al.

RedBench consolidates 37 datasets with 22 risk categories and 19 domains, enabling comprehensive LLM vulnerability evaluation.

Quy-Anh Dang, Chris Ngo, Truong-Son Hy

Systematic analysis of jailbreak mechanisms, evaluation metrics, and unified defense strategies for LLMs and VLMs.

Zejian Chen, Chaozhuo Li, Chao Li et al.

Introduced SiT-Bench, a pure-text spatial reasoning benchmark with 3800+ questions, revealing a significant gap in global consistency for LLMs.

Zhongbin Guo, Zhen Yang, Yushan Li et al.

Proposed STELLA framework combines terminology-aware extraction and synthetic query generation to evaluate aerospace IR models, highlighting semantic understanding with specific results.

Bongmin Kim

Commercial FMCW millimeter-wave radar achieves sub-centimeter water level accuracy with minimal calibration, enabling autonomous monitoring.

Anna Zavei-Boroda, J. Toby Minear, Kyle Harlow et al.