Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Meta-review summarization on PeerSum Research Articles
Loading...
90
Coverage
Automatic decomposition
-3.6
20.7
45
69.3
Jan 27, 2025
Coverage
Faithfulness
Overall Score
Updated 4d ago
Evaluation Results
Method
Method
Links
Coverage
Faithfulness
Overall Score
Automatic decomposition
Backbone=Llama 70B
2025.01
90
90
90
Human-written reference
2025.01
80
80
80
Chunk-wise decomposition
Backbone=Llama 70B
2025.01
70
90
90
Aspect-aware decomposition
Backbone=GPT-4o
2025.01
10
50
50
Sentiment CoT
Backbone=GPT-4o
2025.01
0
0
0
Naive aspect-aware prompting
Backbone=Llama 70B
2025.01
0
0
10
Feedback
Search any
task
Search any
task