PrefBERT, a 150M-parameter reward model trained on human quality ratings, outperforms ROUGE-L and BERTScore as a GRPO reward signal for open-ended long-form generation.
Title resolution pending
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
-
Semantically-Aware Rewards for Open-Ended R1 Training in Free-Form Generation
PrefBERT, a 150M-parameter reward model trained on human quality ratings, outperforms ROUGE-L and BERTScore as a GRPO reward signal for open-ended long-form generation.