Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Scholarly Document Generation on Scholarly papers from 7 venues (AAAI, ACL, CVPR, ICLR, ICML, IEEE, NeurIPS)
Loading...
91.6
ROUGE-L
LLM
90.976
91.138
91.3
91.462
Jun 22, 2026
ROUGE-L
BERTScore F1
ROUGE-L (ref-stripped)
BERTScore (ref-stripped)
Citation Format Accuracy
Citation Recall
Updated 1mo ago
Evaluation Results
Method
Method
Links
ROUGE-L
BERTScore F1
ROUGE-L (ref-stripped)
BERTScore (ref-stripped)
Citation Format Accuracy
Citation Recall
LLM
2026.06
91.6
95.9
91.8
94.1
88
95.6
Dual-Track Framework
2026.06
91
93
89.3
92.2
99.5
91.9
Feedback
Search any
task
Search any
task