Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Text Style Transfer on Treasury style domain
Loading...
50.46
BLEU
Similar 5-shot finetuning
23.6904
30.6402
37.59
44.5398
Feb 16, 2026
BLEU
Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
BLEU
Accuracy
Similar 5-shot finetuning
Backbone=Llama-3.1-8B-...
2026.02
50.46
87.6
Similar 5-shot finetuning w/ terminology and name retrieval
Backbone=Llama-3.1-8B-...
2026.02
50.25
89.4
Similar 3-shot finetuning
Backbone=Llama-3.1-8B-...
2026.02
47.79
82
Random 5-shot finetuning
Backbone=Llama-3.1-8B-...
2026.02
45.22
81.2
Random 3-shot finetuning
Backbone=Llama-3.1-8B-...
2026.02
44.41
79.6
Zero-shot finetuning
Backbone=Llama-3.1-8B-...
2026.02
41.43
82.6
APE with Marian
Backbone=Marian
2026.02
36.37
62.1
5-shot ICL w/ terminology and name retrieval
Backbone=Llama-3.1-8B-...
2026.02
26.69
72.9
5-shot ICL
Backbone=Llama-3.1-8B-...
2026.02
24.72
54.1
Feedback
Search any
task
Search any
task