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.
Learning Tractable Probabilistic Models for Fault Localization
Authors: Aniruddh Nath, Pedro Domingos
AAAI 2016 | Venue PDF | LLM Run Details | Input Tokens: 15,207 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 4,979 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 | We performed an experiment to determine whether TFLM s ability to combine a coverage-based fault localization system with learned bug patterns improves fault localization performance, relative to using the coverage-based system directly. and Results of our experiments are displayed in tables 2 and 3, and ο¬gure 2. |
| Researcher Affiliation | Collaboration | Aniruddh Nath and Pedro Domingos Department of Computer Science & Engineering University of Washington Seattle, WA 98195, U.S.A. EMAIL and Now at Google, Inc. |
| Pseudocode | No | The paper describes procedures and concepts textually but does not include any clearly labeled 'Pseudocode' or 'Algorithm' blocks. |
| Open Source Code | No | The paper mentions 'our unoptimized Python implementation' but does not provide a specific link or an explicit statement about releasing the source code for their methodology. |
| Open Datasets | Yes | We evaluated TFLMs on four mid-sized C programs (table 1) from the Software-artifact Infrastructure Repository (Khurshid et al. 2004). |
| Dataset Splits | Yes | For each subject program, TFLMs were learned via cross-validation, training on all versions of the program except the one being evaluated. and The number of latent subclasses was also chosen via cross-validation, from the range [1, 4]. |
| Hardware Specification | No | The paper states 'our unoptimized Python implementation predicts bug probabilities in a few seconds' but does not specify any particular hardware, such as CPU or GPU models, or memory details, used for the experiments. |
| Software Dependencies | No | The paper mentions using 'SCIKIT-LEARN (Pedregosa et al. 2011) implementation of logistic regression', but it does not provide a specific version number for scikit-learn or any other software dependencies. |
| Experiment Setup | Yes | We use the SCIKIT-LEARN (Pedregosa et al. 2011) implementation of logistic regression, with the class weight= auto parameter, to compensate for the sparsity of the buggy lines relative to bug-free lines. and For TFLMs, we ran hard EM for 100 iterations. and The number of latent subclasses was also chosen via cross-validation, from the range [1, 4]. |