127亿参数模型实现强推理能力，提供可复现的训练方案。
Motif-2-12.7B-Reasoning: A Practitioner's Guide to RL Training Recipes
- 采用两阶段微调+强化学习，用合成数据缓解分布偏移。
- 支持64K上下文，训练稳定性提升，性能媲美更大模型。
- 适合想在有限算力下打造推理模型的研究者和开发者。
我们提出Motif-2-12.7B-Reasoning，一个127亿参数的语言模型，旨在缩小开源系统与闭源前沿模型在复杂推理和长上下文理解上的差距。针对推理适配中常见的模型坍缩与训练不稳定性问题，我们提出一套涵盖系统、数据与算法优化的完整、可复现的训练方案。方法结合了使用混合并行与内核级优化的内存高效基础设施，支持64K token上下文；采用两阶段监督微调（SFT）课程，通过经验证的对齐合成数据缓解分布偏移；此外，详细阐述了基于难度感知数据过滤与多策略轨迹复用的鲁棒强化学习微调（RLFT）流程。实验表明，该模型在数学、编程与代理型基准测试中表现可媲美参数量大得多的模型，在实际算力约束下为社区提供了具备竞争力的开源模型与可扩展的推理能力构建蓝图。
原文摘要 · Abstract (English)
We introduce Motif-2-12.7B-Reasoning, a 12.7B parameter language model designed to bridge the gap between open-weight systems and proprietary frontier models in complex reasoning and long-context understanding. Addressing the common challenges of model collapse and training instability in reasoning adaptation, we propose a comprehensive, reproducible training recipe spanning system, data, and algorithmic optimizations. Our approach combines memory-efficient infrastructure for 64K-token contexts using hybrid parallelism and kernel-level optimizations with a two-stage Supervised Fine-Tuning (SFT) curriculum that mitigates distribution mismatch through verified, aligned synthetic data. Furthermore, we detail a robust Reinforcement Learning Fine-Tuning (RLFT) pipeline that stabilizes training via difficulty-aware data filtering and mixed-policy trajectory reuse. Empirical results demonstrate that Motif-2-12.7B-Reasoning achieves performance comparable to models with significantly larger parameter counts across mathematics, coding, and agentic benchmarks, offering the community a competitive open model and a practical blueprint for scaling reasoning capabilities under realistic compute constraints.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。