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.

This survey categorizes four main methods—latent actions, world models, 2D cues, and 3D reconstructions—for transforming human videos into robot control knowledge, enabling scalable vision-language-action learning.

Zhiyuan Feng, Qixiu Li, Huizhi Liang et al.

HINT-SD employs full-trajectory hindsight analysis for targeted self-distillation, boosting long-horizon RL agent performance by up to 13.60 percentage points with 2.26× training efficiency.

Woongyeong Yeo, Yumin Choi, Taekyung Ki et al.

DADF framework improves watch-time prediction accuracy by 12.57% using distribution-aware debiasing.

Yiqing Yang, Xinlong Zhao, Zhao Liu et al.

URGE algorithm uses Girsanov estimation for unbiased resampling, enhancing diffusion model generation quality.

Chenyang Wang, Weizhong Wang, Yinuo Ren et al.

EXG uses experience graphs to enable self-evolving agents, achieving 150% improvement in task performance and 45.7% efficiency gains.

Yuxin Jin, Siyuan Zhang, Hanchen Wang et al.

This paper derives a n^{-1/4} rate Gaussian approximation for entropy-regularized Q-learning with function approximation, using high-dimensional CLT techniques.

Artemy Rubtsov, Rahul Singh, Eric Moulines et al.

WinQ accelerates low-bit quantization-aware training by 4× using periodic weight interpolation and noise regularization, boosting efficiency and performance.

Dongyue Li, Zechun Liu, Kai Yi et al.

Proposes a constrained optimization attack framework exploiting LLM cascade dependencies, reducing accuracy by 84.6% and inflating costs by 148.9%.

Zehan Sun, Dingfan Chen, Songze Li

A2RBench automates verifiable abstract reasoning benchmark generation, revealing LLMs' poor performance in abstract reasoning.

Qingchuan Ma, Yuexiao Ma, Yongkang Xie et al.

Developed geometric theory of projection heads in self-supervised learning, revealing their role and limitations.

Faris Chaudhry

ECC algorithm calibrates semantic embeddings with model comparisons, improving LLM ranking by 17.64%.

Fangzhou Wu, Sandeep Silwal, Qiuyi Zhang

DeMiAn method enhances robot policy learning with dense language annotations, boosting RoboCasa success by 5 points.

Bosung Kim, Ruiyi Wang, David Acuna et al.

RAGA employs a read-search-verify-construct cycle with knowledge graph-vector synchronization, enabling autonomous KG lifecycle management with evidence provenance.

Chengrui Han, Zesheng Cheng

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.

DRIFT and SAPLMA effectively detect hallucinations after removing benchmark artifacts, achieving AUROC of 0.91.

Khizar Hussain, Murat Kantarcioglu

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.