636 B
636 B
Benchmarks and review files
This directory includes:
500-sample mixed-direction review set
Source-only benchmark for human review:
kashmiri_benchmark_500_source_only.jsonlkashmiri_benchmark_500_stage1_safe_pred.jsonlkashmiri_benchmark_500_stage1_safe_pred.summary.jsonkashmiri_benchmark_500_stage1_safe_pred.xlsxkashmiri_benchmark_500_stage1_safe_pred.summary.xlsx
200-sample ks→en evaluated benchmark
benchmark_simple_ks2en_200.jsonlbenchmark_simple_ks2en_200.stage1_eval.json
The 500-sample file is for manual review. The 200-sample file includes references and saved evaluation output.