SeqGPT: An Out-of-the-Box Large Language Model for Open Domain Sequence Understanding

Authors: Tianyu Yu, Chengyue Jiang, Chao Lou, Shen Huang, Xiaobin Wang, Wei Liu, Jiong Cai, Yangning Li, Yinghui Li, Kewei Tu, Hai-Tao Zheng, Ningyu Zhang, Pengjun Xie, Fei Huang, Yong Jiang

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The experimental results show that Seq GPT has decent classification and extraction ability, and is capable of performing language understanding tasks on unseen domains. We also conduct empirical studies on the scaling of data and model size as well as on the transfer across tasks.
Researcher Affiliation Collaboration Tianyu Yu1*, Chengyue Jiang2*, Chao Lou2*, Shen Huang4* Xiaobin Wang4, Wei Liu2, Jiong Cai2, Yangning Li1, Yinghui Li1, Kewei Tu2 Hai-Tao Zheng1, Ningyu Zhang3, Pengjun Xie4, Fei Huang4, Yong Jiang4 1Tsinghua University 2Shanghai Tech University 3 Zhejiang University 4DAMO Academy, Alibaba Group
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks, nor does it include a figure or section explicitly labeled 'Pseudocode' or 'Algorithm'.
Open Source Code Yes Our models are accessible at https://github.com/Alibaba-NLP/Seq GPT.
Open Datasets Yes We collect and unify 152 datasets across 11 NLU tasks, encompassing not only commonly included information extraction (IE) tasks like NER (Wang et al. 2022a, 2023a)...
Dataset Splits Yes Specifically, the training split of held-in datasets is used during training, no sample from held-out datasets is seen during training, and all tasks involved in held-out datasets are seen during training. For efficiency, we randomly sample 48 records from each evaluation dataset s valid and test split.
Hardware Specification No The paper does not provide specific hardware details such as GPU models, CPU types, or cloud computing instance specifications used for running its experiments. It only generally refers to models and training procedures.
Software Dependencies No The paper does not provide specific software dependency details, such as programming language versions or library versions (e.g., Python 3.8, PyTorch 1.9), that are needed to replicate the experiment.
Experiment Setup No The paper states: 'Most hyperparameters, including optimization steps, learning rates, and batch size, are consistent across all experiments. More training details including hyper-parameters are listed in the appendix to save space.' However, the specific values for these hyperparameters or other training configurations are not included in the provided main text excerpt.