Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates.
FontRL: Chinese Font Synthesis via Deep Reinforcement Learning
Authors: Yitian Liu, Zhouhui Lian2198-2206
AAAI 2021 | Venue PDF | LLM Run Details | Input Tokens: 15,812 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 5,817 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Both quantitative and qualitative experimental results demonstrate the superiority of the proposed Font RL compared to the state of the art. |
| Researcher Affiliation | Academia | Wangxuan Institute of Computer Technology, Peking University, Beijing, P.R. China EMAIL |
| Pseudocode | No | The paper describes its methodology in text and through architectural diagrams (Figure 1 and Figure 2) but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | Yes | Our code is available at https://github.com/lsflytpku/Font RL. |
| Open Datasets | Yes | In our experiments, we directly use the dataset introduced in (Jiang et al. 2019), which consists of glyph images of all 6763 Chinese characters and their manually-specified stroke skeletons in 5 different font styles as target and a mean font style as reference. |
| Dataset Splits | No | The paper mentions using 'an input character set proposed in (Lian, Zhao, and Xiao 2016) for training' consisting of '775 Chinese characters' but does not specify the train/validation/test splits or their proportions for the dataset. |
| Hardware Specification | No | The paper does not provide specific details regarding the hardware used for running experiments, such as CPU or GPU models. |
| Software Dependencies | No | The paper mentions learning rates and hyperparameters but does not provide specific software dependencies with version numbers, such as programming languages or deep learning framework versions. |
| Experiment Setup | Yes | The learning rate of MPNet is initialized as 0.0003, and decayed to 0.0001 after 6000 iterations; the learning rate of BBox Net is initialized as 0.001, decayed to 0.0005 after 40 epochs, and then decayed to 0.0001 after 100 epochs; the hyper-parameters of Style Net are set to the default values. |