The authors unify six extractive content selection tasks under instruction-guided prompting, release a benchmark (IGCS-BENCH) and a 12,490-instance LLM-generated synthetic dataset (GENCS), and show transfer learning from GENCS improves performance on most tasks.
In Proceedings of the 61st Annual Meeting of the Association for Computational Linguis- tics (Volume 1: Long Papers) , pages 9709– 9728, Toronto, Canada
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Unifying Scheme for Extractive Content Selection Tasks
The authors unify six extractive content selection tasks under instruction-guided prompting, release a benchmark (IGCS-BENCH) and a 12,490-instance LLM-generated synthetic dataset (GENCS), and show transfer learning from GENCS improves performance on most tasks.