Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
LLM Decoding on Bitext Telco Gradual Drift
Loading...
0.037
EM
ODD
-0.00148
0.00851
0.0185
0.02849
Feb 8, 2026
EM
Error Distance
BLEU
ROUGE-L
Cosine Similarity
ChrF
BERTScore
Updated 4d ago
Evaluation Results
Method
Method
Links
EM
Error Distance
BLEU
ROUGE-L
Cosine Similarity
ChrF
BERTScore
ODD
Strategy=ODD
2026.02
0.037
0.79
0.663
0.825
0.971
82.824
0.928
Greedy
Strategy=Greedy
2026.02
0
0.702
0.555
0.76
0.855
74.031
0.89
Temp Scaled
Strategy=Temp Scaled
2026.02
0
0.687
0.529
0.745
0.853
72.71
0.885
Feedback
Search any
task
Search any
task