SEGALE extends automatic MT evaluation to long documents, outperforming existing schemes with high correlation to human judgments.

Kuang-Da Wang, Shuoyang Ding, Chao-Han Huck Yang et al.

SPFSplatV2 enables efficient pose-free 3D Gaussian splatting from sparse views, achieving state-of-the-art performance.

Ranran Huang, Krystian Mikolajczyk

Proposes two kernel-based methods for learning Hamiltonian systems, significantly improving data efficiency while preserving energy conservation.

Yasamin Jalalian, Mostafa Samir, Boumediene Hamzi et al.

PTQTP decomposes large model weights into dual ternary planes, enabling multiplication-free inference with near-QLM performance, drastically reducing quantization time.

He Xiao, Runming Yang, Qingyao Yang et al.

Proposed stereo-aware attention and dynamic fusion framework, boosting audio-visual navigation success rate by over 40%.

Jia Li, Yinfeng Yu, Liejun Wang et al.

Proposes Contract-and-Broadcast Self-Attention (CBSA), unifying interpretable and linear-complexity attention via few representatives.

Qishuai Wen, Zhiyuan Huang, Chun-Guang Li

End-to-end LLM-based solver (7B) reduces optimality gap to 1.03-8.20% via SFT and FOARL.

Xia Jiang, Yaoxin Wu, Minshuo Li et al.

Used contrastive learning BERT model to deduplicate Lakh MIDI dataset, filtering out at least 38,134 samples.

Eunjin Choi, Hyerin Kim, Jiwoo Ryu et al.

SCAN: Self-Denoising Monte Carlo Annotation reduces noise in synthetic data, boosting PRM F1 from 19.9 to 59.1 with only 6% inference cost.

Yuyang Ding, Xinyu Shi, Juntao Li et al.

Defines LLM agent generalization via hierarchical domain-task ontology; introduces evaluation metrics and multi-component optimization strategies.

Minxing Zhang, Yi Yang, Roy Xie et al.

By introducing episodic memory, enhance generalization in ML systems, especially for the reversal curse in language models.

Andrew Kyle Lampinen, Martin Engelcke, Yuxuan Li et al.

DiffusionNFT employs flow matching on the forward process, using contrastive positive-negative samples, achieving 25× efficiency without likelihood estimation.

Kaiwen Zheng, Huayu Chen, Haotian Ye et al.

Proposed a self-supervised depth estimation method using mixture distribution to enhance boundary sharpness, achieving a 35% improvement.

Aurélien Cecille, Stefan Duffner, Franck Davoine et al.

LiteLong employs hierarchical topic organization and multi-agent debate for resource-efficient long-text data synthesis.

Junlong Jia, Xing Wu, Chaochen Gao et al.

ScaleCUA leverages cross-platform data to train large-scale GUI agents, achieving 94.4% accuracy on benchmarks.

Zhaoyang Liu, Jingjing Xie, Zichen Ding et al.

TextMineX enhances humanitarian mine action text knowledge extraction accuracy by 44.2% using ontology-guided LLM pipeline.

Chenyue Zhou, Gürkan Solmaz, Flavio Cirillo et al.

Proposed a network flow-based online multi-robot task allocation method incorporating task precedence, coalition effects, and robustness, outperforming offline approaches.

Walker Gosrich, Saurav Agarwal, Kashish Garg et al.

This paper introduces LLM-based automatic modeling and optimization, significantly advancing intelligent decision-making in complex operations research problems.

Yang Wang, Kai Li

GenExam is a multidisciplinary text-to-image benchmark with 1000 samples across 10 subjects, integrating detailed prompts, ground-truth images, and fine-grained scoring.

Zhaokai Wang, Penghao Yin, Xiangyu Zhao et al.

ShinkaEvolve employs parent sampling, novelty rejection, and bandit-based LLM ensemble to achieve sample-efficient open-ended program evolution, outperforming existing methods with only 150 samples.

Robert Tjarko Lange, Yuki Imajuku, Edoardo Cetin