five

skandermoalla/qrpo-paper-llama-sft-magpieair-armorm-temp1-ref50-offpolicy2best-armorm

收藏
Hugging Face2025-12-08 更新2025-12-20 收录
下载链接:
https://hf-mirror.com/datasets/skandermoalla/qrpo-paper-llama-sft-magpieair-armorm-temp1-ref50-offpolicy2best-armorm
下载链接
链接失效反馈
官方服务:
资源简介:
--- license: mit tags: - reinforcement-learning - alignment - qrpo --- # qrpo-paper-llama-sft-magpieair-armorm-temp1-ref50-offpolicy2best-armorm Dataset with reference completions and rewards for a specific model and reward model, ready for training with the **QRPO reference codebase** (https://github.com/CLAIRE-Labo/quantile-reward-policy-optimization). Part of the dataset collection for the paper **Quantile Reward Policy Optimization: Alignment with Pointwise Regression and Exact Partition Functions** (https://arxiv.org/pdf/2507.08068).

许可证:MIT协议 标签: - 强化学习(reinforcement-learning) - 对齐(alignment) - QRPO(Quantile Reward Policy Optimization) # QRPO论文配套Llama监督微调(Supervised Fine-Tuning, SFT)数据集:MagpieAIR-ARMORM(温度系数1、参考样本数50、离线策略2最优版) 本数据集包含特定模型与奖励模型的参考补全结果及奖励评分,可直接用于**QRPO参考代码库**(https://github.com/CLAIRE-Labo/quantile-reward-policy-optimization)的模型训练。 本数据集属于论文**《分位数奖励策略优化:基于逐点回归与精确配分函数的对齐方法》**(https://arxiv.org/pdf/2507.08068)的配套数据集集合之一。
提供机构:
skandermoalla
二维码
社区交流群
二维码
科研交流群
商业服务