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.
SpatialFormer: Semantic and Target Aware Attentions for Few-Shot Learning
Authors: Jinxiang Lai, Siqian Yang, Wenlong Wu, Tao Wu, Guannan Jiang, Xi Wang, Jun Liu, Bin-Bin Gao, Wei Zhang, Yuan Xie, Chengjie Wang
AAAI 2023 | Venue PDF | LLM Run Details | Input Tokens: 17,173 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 5,899 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 | Extensive experiments show that our methods are effective and achieve new state-of-the-art results on few-shot classification benchmarks. [...] Experiment Datasets and Setting [...] Comparing to State-of-the-art Methods Tab. 1 shows the comparison results between our STANet and the related FSL methods on mini Image Net and tiered Image Net [...] Model Analysis |
| Researcher Affiliation | Collaboration | 1 Tencent Youtu Lab, China 2 CATL, China 3 East China Normal University, China 4 Shanghai Jiao Tong University, China |
| Pseudocode | No | The detailed algorithm of STANet is presented in APPENDIX. (The appendix containing the algorithm is not provided in the main paper text.) |
| Open Source Code | Yes | More implementation details of STANet for mini Image Net and tiered Image Net are referred to our released code, such as data augmentation, training epochs, optimizer and learning rate. |
| Open Datasets | Yes | Following (Hou et al. 2019; Xu et al. 2021), two popular benchmark datasets mini Image Net and tiered Image Net are selected, both of which are sampled from Image Net (Krizhevsky, Sutskever, and Hinton 2012) [...] mini Image Net dataset contains 100 categories with 600 images per class, which is separated into {64, 16, 20} categories for {train,validation,test} respectively. tiered Image Net dataset has 608 categories with an average of 1281 images per class, which is divided into {351, 97, 160} categories for {train,validation,test} respectively. |
| Dataset Splits | Yes | mini Image Net dataset contains 100 categories with 600 images per class, which is separated into {64, 16, 20} categories for {train,validation,test} respectively. tiered Image Net dataset has 608 categories with an average of 1281 images per class, which is divided into {351, 97, 160} categories for {train,validation,test} respectively. |
| Hardware Specification | No | The paper does not specify the exact hardware (e.g., GPU/CPU models, memory) used for running the experiments. It only mentions model architectures like Res Net-12 and WRN-28. |
| Software Dependencies | No | The paper does not provide specific version numbers for software dependencies (e.g., Python 3.x, PyTorch 1.x, CUDA x.x). |
| Experiment Setup | Yes | The hyper-parameter λ in Eq. 9 is set to 1.0 according to the results in Tab. 4. More implementation details of STANet for mini Image Net and tiered Image Net are referred to our released code, such as data augmentation, training epochs, optimizer and learning rate. |