Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multilingual Summarization on XL-SUM zh
Loading...
4.78
Token-level Language Confusion (%)
LATB
1.5344
23.4422
45.35
67.2578
Jun 8, 2026
Token-level Language Confusion (%)
Line-level Language Confusion (%)
Response-level Language Confusion (%)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Token-level Language Confusion (%)
Line-level Language Confusion (%)
Response-level Language Confusion (%)
LATB
Backbone=Llama3 8B-I
2026.06
4.78
0
0
Adaptive LATB
Backbone=Llama3 8B-I
2026.06
5.37
0.1
0
Suzume 8B-Multilingual
2026.06
7.56
0.89
0.9
Llama3 8B-I
Prompting Strategy=Str...
2026.06
14.17
9.69
9.1
Llama3 8B-I
Prompting Strategy=Sta...
2026.06
85.92
98.9
98.9
Feedback
Search any
task
Search any
task