Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Text Style Transfer on IRS style domain
Loading...
49.5
BLEU
Similar 5-shot finetuning
26.9216
32.7833
38.645
44.5067
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
49.5
79.6
Similar 5-shot finetuning w/ terminology and name retrieval
Backbone=Llama-3.1-8B-...
2026.02
49.28
89.5
Random 5-shot finetuning
Backbone=Llama-3.1-8B-...
2026.02
48.89
82.6
Similar 3-shot finetuning
Backbone=Llama-3.1-8B-...
2026.02
47.79
74.9
Random 3-shot finetuning
Backbone=Llama-3.1-8B-...
2026.02
47.23
83.9
Zero-shot finetuning
Backbone=Llama-3.1-8B-...
2026.02
42.39
79.3
APE with Marian
Backbone=Marian, RAG m...
2026.02
36.81
64.2
5-shot ICL w/ terminology and name retrieval
Backbone=Llama-3.1-8B-...
2026.02
28.53
67.2
5-shot ICL
Backbone=Llama-3.1-8B-...
2026.02
27.79
59.1
Feedback
Search any
task
Search any
task