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.
Detecting and Measuring Confounding Using Causal Mechanism Shifts
Authors: Abbavaram Gowtham Reddy, Vineeth N Balasubramanian
NeurIPS 2024 | Venue PDF | LLM Run Details | Input Tokens: 27,916 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 5,209 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 | Our empirical results support the usefulness of the proposed measures. |
| Researcher Affiliation | Academia | Abbavaram Gowtham Reddy Indian Institute of Technology Hyderabad EMAIL Vineeth N Balasubramanian Indian Institute of Technology Hyderabad EMAIL |
| Pseudocode | Yes | Algorithm 1: Algorithm for evaluating pairwise CNF-1, CNF-2, CNF-3 |
| Open Source Code | Yes | Code to reproduce the results is presented in the supplementary material. Code is available at https://github.com/gautam0707/CD_CNF. |
| Open Datasets | No | To verify the performance of our method on a large scale, similar to [38], we generate causal graphs of various number nodes using Erdös-Rényi model. |
| Dataset Splits | No | The paper does not explicitly mention training, validation, or test dataset splits. |
| Hardware Specification | No | All the experiments are run on a CPU. |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers. |
| Experiment Setup | Yes | In these experiments, each context is a result of intervention on one node. This is the reason for having the same value for number of nodes N and number of contexts |C|. Sample size denotes the number of data points used in each context. (Table 4 shows N, |C| values and Sample Sizes) |