Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Co-reference Resolution on synthetic dataset
Loading...
81
Accuracy
Oracle
-3.24
18.63
40.5
62.37
Mar 10, 2025
Accuracy
Coverage
Updated 16d ago
Evaluation Results
Method
Method
Links
Accuracy
Coverage
Oracle
Model size (Llama)=3B
2025.03
81
95.38
TokenButler
Model size (Llama)=3B
2025.03
80.2
95.23
Oracle
Model size (Llama)=8B
2025.03
77
93.47
TokenButler
Model size (Llama)=8B
2025.03
76.17
92.59
Oracle
Model size (Llama)=1B
2025.03
49
84.32
TokenButler
Model size (Llama)=1B
2025.03
48.94
84.02
Token Eviction
Model size (Llama)=3B
2025.03
10
51.97
Page-Based
Model size (Llama)=3B
2025.03
6
57.82
Token Eviction
Model size (Llama)=8B
2025.03
3
37.5
Token Eviction
Model size (Llama)=1B
2025.03
1
32.5
Page-Based
Model size (Llama)=1B
2025.03
0
19.78
Page-Based
Model size (Llama)=8B
2025.03
0
46.98
Feedback
Search any
task
Search any
task