TextAttack-Fragile-Interpretations

Code for the paper: "Perturbing Inputs for Fragile Interpretations in Deep Natural Language Processing" (EMNLP BlackboxNLP - 2021)

Pre-calculated candidates and interpretations are available on Google drive here. The results can be replicated by running the results-metric.py script. The exact commmands are detailed in Step-5.

We strongly recommend using conda to manage dependencies.

Run conda create -n frag-exp python=3.6 and subsequently conda activate frag-exp.

Run pip install -r requirements.txt

Following steps re-run the candidate generation process and re-calculate interpretations.

1. Install Textattack from the TextAttack folder's dist folder by installing the wheel: pip install Textattack/dist/textattack-0.2.14-py3-none-any.whl

2. Run python generate_candidates.py --model=distilbert --dataset=sst2 --number=500 --split=validation. All options can be edited for different datasets and models. By default save paths are ./candidates.

3. Run python calculate_interpretations.py --model=distilbert --dataset=sst2 --interpretmethod=IG --number=500 --split=validation. All options can be edited for different datasets and models. By default save paths are ./interpretations.

4. Once all interpretations have been calculated, run python results-metrics.py --model=distilbert --dataset=sst2 --interpretmethod=IG --number=500 --split=validation --metric=rkc.

The available metrics are rkc (Rank Correlation), topk (Top-K Intersection),ppl (Perplexity), grm (Grammar errors) and conf (Model Confidence). Results are stored in ./results.