附录：核心论文完整引用
[1] ReAct: 推理-行动交替范式（基础框架） > Yao, S., Zhao, J., Yu, D., Du, N., Shafran, I., Narasimhan, K., & Cao, Y. (2023). *ReAct: Synergizing Reasoning and Acting in Language Models*. ICLR 2023. > https://openreview.net/forum?id=WE_vluYUL-X
[2] Self-RAG: 自适应检索与自我反思 > Asai, A., Wu, Z., Wang, Y., Sil, A., & Hajishirzi, H. (2023). *Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection*. ICLR 2024. > arXiv:2310.11511
[3] Agentic RAG 综述 > Singh, A., et al. (2025). *Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG*. > arXiv:2501.09136
[4] Deep Research 综述 > Liu, J., et al. (2025). *Deep Research: A Survey of Autonomous Research Agents*. > arXiv:2508.12752v1
[5] DeepSeek-R1: GRPO算法的技术报告 > DeepSeek-AI. (2025). *DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning*. > arXiv:2501.12948
[6] R1-Searcher: 纯RL训练搜索能力 > Song, H., et al. (2025). *R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning*. > arXiv:2503.05592
[7] DeepResearcher: 真实环境RL训练 > Zheng, Y., et al. (2025). *DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-World Environments*. > arXiv:2504.03160
[8] Process vs Outcome Reward: Agentic RAG奖励设计 > Zhang, W., et al. (2025). *Process vs. Outcome Reward: Which is Better for Agentic RAG Reinforcement Learning*. > arXiv:2505.14069v1
[9] Auto-RAG: 自主检索增强生成 > Yu, T., Zhang, S., & Feng, Y. (2024). *Auto-RAG: Autonomous Retrieval-Augmented Generation for Large Language Models*. > arXiv:2411.19443
[10] ZeroSearch: 无需搜索数据训练搜索能力 > Sun, H., et al. (2025). *ZeroSearch: Incentivize the Search Capability of LLMs without Searching*. > arXiv:2505.04588
---
视频展示顺序建议（每篇停留3-4秒）： 1. ReAct → 2. Self-RAG → 3. Agentic RAG Survey → 4. Deep Research Survey → 5. DeepSeek-R1 → 6. R1-Searcher → 7. DeepResearcher → 8. Process vs Outcome Reward → 9. Auto-RAG → 10. ZeroSearch
#深度研究 #论文引用 #AgenticRAG #小凯