braindatalab/gecobench

Issues

• Incomplete gender perturbations in the subject dataset

  #84 opened 2 months ago
  1

• We usw the right latex template

  #67 opened 5 months ago
  0

• Add prediction for model evaluation

  #63 opened 6 months ago
  0

• Bug: No optimizer.zero_grad in training function

  #62 opened 7 months ago
  0

• Research bias metrics for classification models

  #61 opened 5 months ago
  1

• Decide which dataset to use for gender (subsampling / no subsampling)

  #60 opened 6 months ago
  1

• Research bias metrics for datasets only

  #59 opened 5 months ago
  0

• Possible normalization strategy for matrix sums of co-occurence metric

  #58 opened 5 months ago
  2

• Research applicable classification bias metric for models

  #55 opened 5 months ago
  1

• Data (train, test.jsonl) does not contain both gender sentences

  #54 opened 7 months ago
  3

• Location of the dataframe.pkl files needed for generate_data.py

  #49 opened 7 months ago
  1

• Provisioning of our datasets

  #48 opened 7 months ago
  1

• We have a normalization strategy for computing word-based most common feature attribution

  #47 opened 7 months ago
  0

• Bug: Gradient Shap

  #45 opened 7 months ago
  0

• Add comments in the code

  #43 opened 7 months ago
  0

• Add explanation in README on how to use poetry

  #42 opened 5 months ago
  0

• Implement Bias Metrics (DisCO, LPBS)

  #38 opened 7 months ago
  0

• Visualization of the whole project

  #31 opened 5 months ago
  0

• Visualization of data generation process (LaTeX table)

  #30 opened 7 months ago
  0

• Bug: correct_classified_intersection

  #28 opened 6 months ago
  0

• Restructure dataset config

  #27 opened 9 months ago
  1

• Check sentiment output and start sentiment train run

  #26 opened 9 months ago
  0

• Investigate dataset from Cabello, Laura, et al. paper

  #23 opened 7 months ago
  0

• Literature research on biases in LLMs

  #22 opened 7 months ago
  0

• Investigate concept of bias in Cabello, Laura, et al. paper

  #21 opened 7 months ago
  0

• Visualization of explanations from multiple pre-trained models in one plot

  #20 opened 10 months ago
  0

• Visualize correlation between words and labels

  #18 opened 10 months ago
  0

• Visualize sentiment classifier on gender dataset

  #16 opened 7 months ago
  0

• Add sentiment datasets to pipeline

  #15 opened 9 months ago
  0

• Text dataset have .jsonl or .csv format

  #14 opened 8 months ago
  0

• Simple attention model

  #13 opened 7 months ago
  0

• Bert attention as explanations

  #12 opened 5 months ago
  0

• Data and model storage

  #11 opened 5 months ago
  1

• Correlation between words and label as 'XAI method'

  #10 opened 7 months ago
  0

• Why do DeepLift, Guided BackProp and DeepLift produce similar explanation performances?

  #9 opened 5 months ago
  0