Talk Funny! A Large-Scale Humor Response Dataset with Chain-of-Humor Interpretation
Authors: Yuyan Chen, Yichen Yuan, Panjun Liu, Dayiheng Liu, Qinghao Guan, Mengfei Guo, Haiming Peng, Bang Liu, Zhixu Li, Yanghua Xiao
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive evaluations demonstrate that our proposed dataset and auxiliary tasks effectively help PLMs to generate humorous responses, laying the groundwork for future humor research. |
| Researcher Affiliation | Collaboration | 1Shanghai Key Laboratory of Data Science, School of Computer Science, Fudan University 2Institute of Automation, Chinese Academy of Sciences 3School of Computer Science, Beijing Institute of Technology 4Alibaba DAMO Academy 5University of Zurich 6 Beijing Jiaotong University, 7 RALI & Mila, Universit e de Montr eal 8Fudan-Aishu Cognitive Intelligence Joint Research Center, Shanghai, China |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide concrete access to source code for the methodology described, nor does it explicitly state that its code is open-source or available. |
| Open Datasets | No | The paper states, "we collect and annotate 4,116 pieces of explainable humorous context-response pairs with statistics shown in Table 1.", but does not provide a specific link, DOI, repository name, or citation for public access to the constructed dataset. |
| Dataset Splits | No | The paper mentions "use early stopping with 50 epochs" which implies a validation set, but it does not provide specific dataset split information (percentages, sample counts, or clear predefined splits) for training, validation, and testing. |
| Hardware Specification | Yes | The experiment is carried out on one Tesla V100 GPUs with Pytorch in Python. |
| Software Dependencies | No | The paper mentions "Pytorch in Python" but does not provide specific version numbers for these software components or any other libraries used. |
| Experiment Setup | Yes | For the annotation-enhanced humor response, the maximum source length and targeted length is set to 512 and 128, respectively. We initialize the learning rate from 2e-5 to 4e-5 and batch size to 8 according to the memory of the machine, and use early stopping with 50 epochs. |