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.
Causal Representation Learning via Counterfactual Intervention
Authors: Xiutian Li, Siqi Sun, Rui Feng
AAAI 2024 | Venue PDF | LLM Run Details | Input Tokens: 18,694 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 4,220 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 | Through extensive experiments on realworld and synthetic datasets, we show that our method outperforms different baselines and obtains the state-of-the-art results for achieving causal representation learning. |
| Researcher Affiliation | Academia | 1School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing, Fudan University, Shanghai 200433 2Fudan Zhangjiang Institute, Shanghai, 200120 3Shanghai Collaborative Innovation Center of Intelligent Visual Computing |
| Pseudocode | No | The paper does not contain any sections or figures explicitly labeled as 'Pseudocode' or 'Algorithm'. |
| Open Source Code | No | The paper does not contain any statement about releasing source code or provide any links to a code repository for the methodology described. |
| Open Datasets | Yes | 1) Celeb A dataset contains face images with 40 attributes annotations. Following previous causally disentangling methods (Yang et al. 2021; Shen et al. 2022), we select images with different attributes subsets as Age dataset and Smile dataset respectively. ... 2) For synthetic datasets, we build Pendulum (Pend) dataset and Flow dataset like (Yang et al. 2021). |
| Dataset Splits | No | For both Age and Smile datasets, we randomly select 30,000 images for training and 7,500 images for test. ... For both Pend and Flow, we build 5,000 images for training and 1,000 images for test. |
| Hardware Specification | No | The paper does not specify the hardware used for running experiments, such as GPU or CPU models. |
| Software Dependencies | No | The paper mentions optimizers like Adam and Cosine Annealing but does not provide specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions). |
| Experiment Setup | Yes | During training, batch size is set as 64. For optimizing, we utilize Adam (Kingma and Ba 2014) with Cosine Annealing (Loshchilov and Hutter 2016). |