Benchmark and Code for ELI-Why
The ELI-Why benchmark is stored as ELI_Why.jsonl.
- RQ1/
- Individual Metrics/
Contains scripts for running individual metrics and analyses. To run individual metrics, navigate to the
RQ1/directory and execute:
python3 individual_metric_readability_analysis.py --model GPT4o --set fullAlternatively, to run all combinations, use:
python3 individual_metric_readability_analysis.py --all- TESDIFF(Similarity) Metric/
Contains scripts for running TESDIFF metric and analyses. To run it, in the
RQ1/directory and execute:
python3 tesdiff.py --all