CNNSum adds a 16k-128k Chinese novel summarization benchmark with human annotations and evidence that short-context fine-tuning, base models, and RoPE-base scaling improve long-context summarization.
In Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence , pages 8299– 8307
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
CNNSum: Exploring Long-Context Summarization with Large Language Models in Chinese Novels
CNNSum adds a 16k-128k Chinese novel summarization benchmark with human annotations and evidence that short-context fine-tuning, base models, and RoPE-base scaling improve long-context summarization.