Learning to Reason by Analogy via Retrieval-Augmented Reinforcement Fine-Tuning

**摘要**
Retrieval-augmented generation (RAG) has become a standard mechanism for grounding language models in external knowledge, yet conventional retrieval based on lexical or semantic similarity is poorly suited for complex reasoning tasks: a semantically similar problem may demand an entirely different solution strategy, while a superficially different problem may share the same underlying reasoning pa
👤 作者: Zilin Xiao, Qi Ma, Chun-cheng Jason Chen, Xintao Chen, Avinash Atreya, Hanjie Chen, Vicente Ordonez

---
🔗 **[Learning to Reason by Analogy via Retrieval-Augmented Reinforcement Fine-Tuning](https://arxiv.org/abs/2606.13680v1)**

> Learning to Reason by Analogy via Retrieval-Augmented Reinforcement Fine-Tuning
🏷️ 来源: ArXiv cs.AI
⏱️ 2026-06-12 14:00

Learning to Reason by Analogy via Retrieval-Augmented Reinforcement Fine-Tuning

回复