CRYSTAL benchmark evaluates multimodal reasoning transparency using Match F1 and Ordered Match F1, revealing systematic flaws in existing models.

Wayner Barrios, SouYoung Jin

SldprtNet is a large-scale multimodal dataset with 242,000 industrial parts for semantic-driven CAD modeling.

Ruogu Li, Sikai Li, Yao Mu et al.

Zero-hyperparameter multi-corner analysis using learned priors reduces validation cost by over 10 times.

Wei W. Xing, Kaiqi Huang, Jiazhan Liu et al.

Influence Malleability in Linearized Attention: Dual Implications of Non-Convergent NTK Dynamics.

Jose Marie Antonio Miñoza, Paulo Mario P. Medina, Sebastian C. Ibañez

InterEdit uses Semantic-Aware Plan Token Alignment and Interaction-Aware Frequency Token Alignment for multi-human 3D motion editing.

Yebin Yang, Di Wen, Lei Qi et al.

DDIM reverse chain as Partitioned Iterated Function Systems provides a unified design language for denoising diffusion models.

Ann Dooms

WALAR method enhances low-resource language translation using monolingual data, surpassing LLaMAX model.

Yifeng Liu, Siqi Ouyang, Yatish Hosmane Revanasiddappa et al.

Proposed a PCA sweep method to optimize dimension selection in SSD, enhancing interpretability and stability.

Hubert Plisiecki, Maria Leniarska, Jan Piotrowski et al.

Federated few-shot learning on neuromorphic hardware using FedUnion strategy achieves 77.0% accuracy.

Steven Motta, Gioele Nanni

OpenSWE synthesizes 45,320 executable Docker environments via a multi-agent pipeline, enhancing SWE agent training efficiency.

Dayuan Fu, Shenyu Wu, Yunze Wu et al.

Structured distillation reduces personalized agent memory tokens by 11x while preserving retrieval capabilities.

Sydney Lewis

Significantly reduces manipulator end-effector deviation under cyberattacks using a novel active defense method.

Gabriele Gualandi, Alessandro V. Papadopoulos

Through resource-centric route fragmentation, improve multi-robot path planning efficiency in agricultural environments, achieving 95% task throughput.

James R. Heselden, Gautham P. Das

Long-form RewardBench evaluates reward models for long-form generation, revealing current models' deficiencies in long-form reward modeling.

Hui Huang, Yancheng He, Wei Liu et al.

RoboStream employs Spatio-Temporal Fusion Tokens and Causal Spatio-Temporal Graphs to enable persistent memory and reasoning, achieving 90.5% success in long-horizon robotic tasks.

Yuzhi Huang, Jie Wu, Weijue Bu et al.

MotionAnymesh transforms static 3D meshes into simulation-ready digital twins using physics-constrained joint estimation and motion optimization.

WenBo Xu, Liu Liu, Li Zhang et al.

AMRO-S uses Ant Colony Optimization for efficient, interpretable multi-agent LLM routing, achieving 4.7x speedup.

Xudong Wang, Chaoning Zhang, Jiaquan Zhang et al.

HMS-BERT uses hybrid multi-task self-training for multilingual, multi-label cyberbullying detection, achieving a macro F1-score of 0.9847.

Zixin Feng, Xinying Cui, Yifan Sun et al.

Study the mechanism of diffusion models learning data statistics from simple to complex using the mixed cumulant model.

Lorenzo Bardone, Claudia Merger, Sebastian Goldt

Introduces Hessian Relative Uniform Continuity (HRUC) to simplify distributed mirror descent analysis, covering a broader range of kernels.

Junwen Qiu, Ziyang Zeng, Leilei Mei et al.