arXiv subjects

Wang Qun

Publications and source records attributed to Wang Qun.

3 recordsLinked to original sources

arXiv2024

Xmodel-2 Technical Report

Xmodel-2 is a 1.2-billion-parameter large language model designed specifically for reasoning tasks. Its architecture enables different model scales to share a unified set of hyperparameters, allowing for extensive experimentation on smaller models and seamless transfer of optimal configurations to larger models. To maximize training efficiency and stability, Xmodel-2 employs the WSD learning rate scheduler from MiniCPM. Pretrained on 1.5 trillion tokens from diverse sources, Xmodel-2 achieves state-of-the-art performance in complex reasoning and agent-based tasks, while maintaining low training costs. These results highlight the potential of efficient model design and training strategies in advancing reasoning capabilities. Model checkpoints and code are publicly available on GitHub at https://github.com/XiaoduoAILab/Xmodel-2

cs.AI↗

arXiv2024

Xmodel-1.5: An 1B-scale Multilingual LLM

We introduce Xmodel-1.5, a 1-billion-parameter multilingual large language model pretrained on 2 trillion tokens, designed for balanced performance and scalability. Unlike most large models that use the BPE tokenizer, Xmodel-1.5 employs a custom unigram tokenizer with 65,280 tokens, optimizing both efficiency and accuracy. The model delivers competitive results across multiple languages, including Thai, Arabic, French, Chinese, and English, outperforming Alibaba's PolyLM-1.7B on respective evaluation datasets. Xmodel-1.5 excels in benchmarks like mMMLU and PIQA, and achieves state-of-the-art results in Thai. To support low-resource language research, we release Xdata_Thai, a Thai-specific evaluation dataset featuring unique linguistic challenges such as gendered particles and idioms. While the model demonstrates strong performance, there is still room for improvement in handling culturally specific nuances. We hope this work contributes to advancements in multilingual AI research. Models and code are publicly available on GitHub at https://github.com/XiaoduoAILab/XmodelLM-1.5

cs.CL↗

arXiv2022

Existence and Uniqueness of Reynolds equation under natural boundary conditions

For the problem of solving Reynolds equation under natural boundary conditions, the corresponding hypothetical solution can be obtained by assuming the free boundary. If the solution satisfies natural boundary conditions, then the boundary is the boundary we are looking for. Obviously, there is a set S formed by all the boundaries that assume the solution is positive. We prove equivalence between maximum element of S and natural condition. We prove the closeness of set S under addition. Therefore, we prove that the set S must have a unique greatest element. Furthermore, we obtain the uniqueness and existence of solutions of Reynolds equation under natural boundary conditions. In engineering, the zero setting method is often used to find the boundary of Reynolds equation, and we also give the proof that the zero setting method has only one iterative error solution. We give an algorithm for solving Reynolds equation under one-dimensional conditions, which reaches the theoretical upper bound. We discuss the physical meaning of this method at the end.

math.AP↗