2 code implementations • 12 Feb 2024 • Yifan Zhang, Yifan Luo, Yang Yuan, Andrew Chi-Chih Yao
Our method showcases a 2 times increase in pretraining token efficiency compared to state-of-the-art baselines, underscoring the potential of our approach in enhancing models' mathematical reasoning capabilities.
1 code implementation • 17 Jan 2024 • Haoxiong Liu, Yifan Zhang, Yifan Luo, Andrew Chi-Chih Yao
The MMIQC dataset is available on the HuggingFace hub at https://huggingface. co/datasets/Vivacem/MMIQC.
Ranked #41 on Math Word Problem Solving on MATH (using extra training data)
no code implementations • 22 Oct 2023 • Yifan Luo, Yiming Tang, Chengfeng Shen, Zhennan Zhou, Bin Dong
In this paper, we propose an optimal control framework tailored for multi-round interactions with LLMs.
1 code implementation • 5 Jul 2023 • Shengding Hu, Yifan Luo, Huadong Wang, Xingyi Cheng, Zhiyuan Liu, Maosong Sun
In this paper, we find that the PLMs already possess the knowledge required to rebut such questions, and the key is how to activate the knowledge.
no code implementations • 25 May 2023 • Yifan Luo, Bin Dong
In this paper, we studied two identically-trained neural networks (i. e. networks with the same architecture, trained on the same dataset using the same algorithm, but with different initialization) and found that their outputs discrepancy on the training dataset exhibits a "double descent" phenomenon.
2 code implementations • 21 Jan 2021 • Yunpeng Gong, Zhiyong Zeng, Liwen Chen, Yifan Luo, Bin Weng, Feng Ye
This method can not only improve the accuracy of the model, but also help the model defend against adversarial examples; 2) Multi-Modal Defense, it integrates three homogeneous modal images of visible, grayscale and sketch, and further strengthens the defense ability of the model.
Ranked #19 on Person Re-Identification on Market-1501-C
no code implementations • 18 Oct 2020 • Yifan Luo, Jindan Xu, Wei Xu, Kezhi Wang
Federated learning (FL) in a bandwidth-limited network with energy-limited user equipments (UEs) is under-explored.