OpenGenAlign: A Preference Dataset and Benchmark for Trustworthy Reward Modeling in Open-Ended, Long-Context Generation

Documenti analoghi

• DuaShepherd: Integrating Stepwise Correctness and Potential Rewards for Mathematical Reasoning
  di: Wu, Yuanhao, et al.
  Pubblicazione: (2025)
• RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Models
  di: Niu, Cheng, et al.
  Pubblicazione: (2023)
• From General to Targeted Rewards: Surpassing GPT-4 in Open-Ended Long-Context Generation
  di: Guo, Zhihan, et al.
  Pubblicazione: (2025)
• VeraCT Scan: Retrieval-Augmented Fake News Detection with Justifiable Reasoning
  di: Niu, Cheng, et al.
  Pubblicazione: (2024)
• Tournament-GRPO: Group-Wise Tournament Rewards for Reinforcement Learning in Open-Ended Long-Form Generation
  di: Yang, Zixuan, et al.
  Pubblicazione: (2026)