SharpMoE leverages predicted clean latent features for saliency-aware routing, boosting diffusion MoE image generation by ~20% in FID.

Haoyou Deng, Keyu Yan, Chaojie Mao et al.

PortraitGen injects real exemplars into GRPO and combines OmniReward with AI-Portrait, reaching 96.35% artifact-detection accuracy.

Xiaomin Li, Qian Liang, Yinan Li et al.

AgentX employs a multi-agent system for autonomous self-iteration in industrial recommender systems, significantly boosting efficiency and business value.

Changxin Lao, Fei Pan, Guozhuang Ma et al.

PlanRL integrates Frenet-based polynomial trajectory planning with RL, achieving 5% score improvement and 19% success rate increase on CARLA benchmarks.

Joonhee Lim, Yongjae Lee, Jangho Shin et al.

Proposes FL-based quantization methods reducing communication and computation costs, enhancing robustness and privacy.

Farwa Ikram, Dipanwita Thakur, Antonella Guzzo et al.

Proposes AHOIS, a multi-agent system using Socratic questioning for autonomous discovery in high-dimensional physical systems, validated on multimode fiber platform.

Xianrui Zeng, Pengfei Liu, Yirui Zang et al.

Proposes Bézier Walk Evolution (BWE), integrating geometry-driven path construction with distance-aware random walks, balancing exploration and exploitation via adaptive curve order.

Jinpeng Wang, Xingguo Xu, Yujing Sun et al.

PhysEditWorld, a UE5-based multimodal dataset, enables explicit control of physics (gravity) via matched replays, supporting physics-aware world modeling.

Bin Hu, Yanwen Ma, Jiehui Huang et al.

TGHE: Template-based Graph Homomorphic Encryption enables scalable privacy-preserving GNN inference on large dynamic graphs, achieving 66.9× speedup with minimal accuracy loss.

Ngoc Bao Anh Le, Thai T. Vu, John Le et al.

CAT-Q enables ultra-efficient ternary quantization of large language models using only 512 calibration samples, outperforming state-of-the-art QAT methods on models up to 235B parameters.

Shigeng Wang, Chao Li, Yangyuxuan Kang et al.

EcoTable uses graph models and LLMs to enable cost-effective, natural language-driven table integration, enhancing data lake query support.

Yuhui Wang, Jinqi Liu, Chengliang Chai et al.

DiCoBench introduces a high-resolution multi-image benchmark focusing on differential and commonality visual cues, revealing models' significant gaps in fine-grained perception, with top accuracy at 58.1%.

Geng Li, Yuxin Peng

Introduces PhyEditBench, a real-world video-based benchmark for physics reasoning, revealing current models' limitations in physical understanding.

Shengbin Guo, Shaokang He, Chaoyue Meng et al.

Play2Perfect uses play pretraining on diverse objects to achieve precise assembly, improving sample efficiency by 33x.

Tyler Ga Wei Lum, Kushal Kedia, C. Karen Liu et al.

Introduces a preference-conditioned Bellman operator using Chebyshev scalarization for deterministic Pareto-optimal policies in MOMDPs, ensuring full Pareto front coverage.

Aniruddha Joshi, Niklas Lauffer, Sanjit Seshia

Developed a multi-QPU-compatible distributed QAOA simulator for large-scale QUBO problems in engineering, improving efficiency and scalability.

Ali Rajabi, Milad Hasanzadeh, Amin Kargarian

Introduces Red Queen Gödel Machine (RQGM), a co-evolving framework with learned evaluators, achieving 1.35-1.86× efficiency gains in code, science, and proof tasks under non-stationary utilities.

Alex Iacob, Andrej Jovanović, William F. Shen et al.

SimPhysNet combines physics-informed neural networks with self-supervised contrastive learning to achieve 96.06% accuracy in laser welding penetration prediction using only 200 labeled images.

Sen Li, Xiaoying Liu, Xiaojian Xu et al.

This paper develops a theoretical framework for synthetic augmentation in score-based imbalanced classification, showing limited gains under well-specified models and potential improvements under model misspecification.

Zhengchi Ma, Pengfei Lyu, Anru R. Zhang

This study reveals the asymmetric dynamics of rule survival in pretraining models, with support frequency predicting rule fate and demonstrating irreversibility via editing interventions.

Juliana Li, Diya Sreedhar