Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-Doc Question Answering on LongBench-E
Loading...
49.6
F1 Score
Teacher Model (w/ Context)
5.504
16.952
28.4
39.848
Oct 23, 2025
F1 Score
Updated 22d ago
Evaluation Results
Method
Method
Links
F1 Score
Teacher Model (w/ Context)
Backbone=Qwen3-1.7B, C...
2025.10
49.6
Mean-Pooling
Backbone=Qwen3-1.7B, C...
2025.10
45.9
Compression-Tokens (Bidirectional)
Backbone=Qwen3-1.7B, C...
2025.10
43
Mean-Pooling
Backbone=Qwen3-1.7B, C...
2025.10
41.4
Compression-Tokens (Causal)
Backbone=Qwen3-1.7B, C...
2025.10
40.9
Compression-Tokens (Bidirectional)
Backbone=Qwen3-1.7B, C...
2025.10
38.2
Mean-Pooling
Backbone=Qwen3-1.7B, C...
2025.10
36
Compression-Tokens (Causal)
Backbone=Qwen3-1.7B, C...
2025.10
32.5
Compression-Tokens (Bidirectional)
Backbone=Qwen3-1.7B, C...
2025.10
32.1
Compression-Tokens (Causal)
Backbone=Qwen3-1.7B, C...
2025.10
31.6
LLMLingua2
Backbone=Qwen3-1.7B, C...
2025.10
29.4
PCC Large
Backbone=Llama3.1-8B,...
2025.10
28
LLMLingua2
Backbone=Qwen3-1.7B, C...
2025.10
22.4
LLMLingua2
Backbone=Qwen3-1.7B, C...
2025.10
21.8
Teacher Model (w/o Context)
Backbone=Qwen3-1.7B, C...
2025.10
21.7
PCC Large
Backbone=Llama3.1-8B,...
2025.10
11.4
PCC Large
Backbone=Llama3.1-8B,...
2025.10
7.2
Feedback
Search any
task
Search any
task