总结

本书完整梳理了从 Transformer 基础、面向对齐的强化学习，到自主智能体系统构建的全过程。跨越各章节浮现出的关键主题如下：

  1. 对齐是一个系统工程问题。仅有一个好的损失函数并不够。生产级 RLHF 需要同时管理 4 个以上的模型、在数百块 GPU 上分布计算、处理容错，并监控奖励作弊行为——而这一切必须同时进行。
  2. 不存在单一最优方法。PPO 仍是追求极致质量的金标准，但需要巨大的工程投入。DPO 及其变体则为基础设施有限的团队提供了极具吸引力的折中。GRPO 弥合了可验证奖励领域的鸿沟。正确的选择取决于你的数据、算力预算与质量门槛。
  3. 推理能力可以从奖励中涌现。DeepSeek-R1 证明，思维链、自我验证与回溯能力可以从简单的二值奖励信号与组相对优化中涌现——无需对推理过程进行显式示范。测试时算力扩展意味着「更小的模型 + 更多思考」可以匹敌更大的模型。
  4. 标准开启生态。MCP 将工具集成问题从 N×M 降到 N+M。A2A 让不同团队构建的智能体在不共享内部实现的前提下协作。这些协议之于智能体 AI，正如 HTTP 之于 Web——是开放生态的基础设施。
  5. 智能体是自然的下一步。一旦模型被对齐，前沿就从「单次回复有多好？」转向「模型能否自主解决多步问题？」这需要新的训练范式（带环境奖励的智能体 RL）、新的基础设施（运行框架、工具协议、记忆系统）以及新的评估方法（轨迹级基准）。
  6. 评估驱动一切。没有严格的评估——从奖励模型验证到智能体任务成功率，从污染检测到 LLM-as-Judge 校准——进步无法度量，回退也不可见。你所选择的基准塑造了你所构建的系统。
  7. 简单才能扩展。最可靠的生产级智能体采用满足需求的最简单架构——优先使用 prompt chaining 与 routing，而非自主循环；优先使用单智能体，而非多智能体集群。复杂度只能通过被证实的需要去赢得。

前路：开放挑战

从交互中学习

当前的 RLHF 流水线 [280] 将对齐视为一次性的训练阶段。未来则指向 从部署中持续学习：智能体能够从每一次用户交互、工具失败与环境观察中改进——同时避免灾难性遗忘（Catastrophic Forgetting）[186] 与奖励漂移。关键开放问题：

  • 非平稳奖励分布下的在线学习。
  • 生产环境中的安全探索 [107]（在学习过程中避免有害动作）。
  • 在长智能体轨迹（数百次工具调用）上的高效信用分配。

可扩展监督

当智能体的能力越来越强时，人类监督本身成为瓶颈。当前方法（RLHF [280]、Constitutional AI [16]）依赖人类评估模型输出——但当模型输出超出人类理解能力时会发生什么？

  • 递归奖励建模[57]：用 AI 协助人类评估 AI。
  • 辩论与放大[157]：两个模型相互辩论，由人类裁判哪一方的论证更令人信服。
  • 基于过程的监督[216]：奖励正确的推理步骤，而不仅是最终答案。
  • 机制可解释性[273]：理解模型内部 在做什么，而不仅是它输出了什么。

世界模型与规划

当前智能体是反应式的——它们一次只观察并响应一步。未来智能体将需要 内部世界模型[130] 以支持前瞻式规划：

  • 在执行动作之前预测其后果。
  • 对可能动作序列进行树搜索（类似 AlphaGo [331] 与 MuZero [316]，但面向开放任务）。
  • 从交互轨迹中学习环境的状态转移。

多智能体生态

A2A 协议 [115] 与多智能体框架预示着这样一种未来：成百上千的专用智能体协作、协商并相互委派——形成所谓的「智能体经济（Economy of Agents）」[267]。开放挑战包括：

  • 服务于不同委托方的智能体之间的信任与验证。
  • 竞争性场景中涌现的合作与涌现的欺骗之间的对比 [153]。
  • 资源分配（算力、工具访问、优先级）的市场机制。
  • 治理：当一条由 10 个智能体组成的链路产生有害结果时，由谁负责？[6]

智能体安全与信任

自主智能体继承了其底层 LLM 的全部安全漏洞，再加上工具访问、多智能体委派与持久化记忆所带来的新攻击面（第 19–21 章）。关键的未解问题：

  • 规模化的 Prompt 注入[120]：随着智能体消费不可信内容（网页、邮件、API 响应），间接 prompt 注入成为系统性问题。目前尚无稳健的防御手段。
  • 混淆代理攻击：持有合法凭证的智能体可能被骗，代表潜伏在数据流中的攻击者 [10] 滥用其凭证。
  • 不削弱能力的沙箱化：最小权限执行限制了智能体能做的事，但过度严格的沙箱会抹杀智能体的价值。如何找到合适的边界仍是开放的设计问题。
  • 审计与归因：当智能体链跨越多个组织（通过 A2A [115]）时，追溯 谁 授权了 什么 动作在架构上仍未解决。
  • 信任校准：智能体必须学会何时 不要 信任——工具响应是否真实、其他智能体的声明是否经过验证。

超越基准的评估

第 14 章揭示了基准塑造了我们所构建的系统——然而当前评估仍存在关键缺口：

  • 真实部署指标：SWE-bench [169] 与 GAIA [254] 等基准衡量的是孤立任务；而生产级智能体面对的是模糊目标、变动的需求与多轮恢复。
  • 奖励模型有效性：RLHF 假设奖励模型能捕捉人类偏好，但奖励作弊 [333] 与分布漂移会在规模化时侵蚀这一假设。
  • 成本—质量前沿：两个智能体可能取得相同准确率，但其中一个的 token 成本高出 10×。评估必须开始关注成本。
  • 分布漂移下的安全性：在测试中安全的智能体在新输入下可能行为不安全。智能体规模上的对抗评估 [290] 与红队测试仍不成熟。

效率与可获得性

对 70B 模型做 RLHF 训练成本为 10​K−−100K。运行自主智能体在每个复杂任务上的成本为 1−−50。要让智能体 AI 产生广泛影响，需要：

  • 将智能体能力从大模型蒸馏到小模型 [139, 184]。
  • 更高效的 RL 算法（更少样本、更低方差）[319]。
  • 无需云端往返即可运行的端侧智能体。
  • 在智能体任务上达到专有模型质量的开源权重模型 [69]。

延伸阅读

基础论文

  • Attention Is All You Need[357]——Transformer 架构。
  • RLHF / InstructGPT[280]——首个大规模 RLHF 部署。
  • PPO[319]——近端策略优化（Proximal Policy Optimization）。
  • DPO[302]——直接偏好优化（Direct Preference Optimization）。
  • GRPO / DeepSeek-R1[323, 69]——组相对策略优化与涌现推理。
  • ReAct[409]——面向 LLM 智能体的推理 + 行动框架。
  • Toolformer[315]——教会 LLM 使用工具。
  • RAG[209]——检索增强生成。

系统与扩展

  • Megatron-LM[327]——张量并行与流水线并行。
  • DeepSpeed ZeRO[304]——内存高效的分布式训练。
  • vLLM[196]——用于高效 LLM 服务的 PagedAttention。
  • Flash Attention[64]——IO 感知的精确注意力。

智能体 AI

  • Building Effective Agents[9]——设计模式与原则。
  • Voyager[362]——在 Minecraft 中拥有技能库的开放式智能体。
  • SWE-bench[169]——面向自主软件工程的基准。
  • OSWorld[397]——完整的计算机使用基准。
  • GAIA[254]——面向真实世界任务的通用 AI 助手基准。
  • MemGPT[281]——受操作系统启发的无界上下文记忆管理。
  • Model Context Protocol[10]——工具集成的开放标准。
  • Agent-to-Agent Protocol[115]——智能体间通信标准。

对齐与安全

  • Constitutional AI[16]——自监督对齐。
  • Sleeper Agents[153]——欺骗性对齐隐患。
  • Reflexion[326]——从语言化自我反思中学习。
  • Indirect Prompt Injection[120]——LLM 集成应用的安全风险。

在线资源免责声明本文档是作者独立撰写的综述与教育资源。文中表达的观点、意见与结论仅代表作者本人，并不必然反映作者目前或曾经服务的任何雇主、组织或机构的立场。本作品不包含任何专有、机密或商业秘密信息。所有引用材料均取自公开来源，包括同行评审出版物、开放获取的预印本、官方文档以及开源代码仓库。本内容按“原样”提供，不附带任何明示或暗示的保证。作者不就信息对任何特定用途的准确性、完整性或适用性作出任何陈述。读者在将任何论断、公式或实现细节应用于生产系统之前，应当独立核实。欢 迎 反 馈。 如 果 您 发 现 任 何 错 误、 不 准 确 之 处 或 有 改 进 建 议， 欢 迎 将 反 馈 发 送 至[first_name]r@gmail.com。AI 使用披露。本书使用大语言模型作为研究与起草辅助工具。所有内容均由作者尽最大努力进行了编辑与核实。许可证。本作品采用 Creative Commons Attribution-ShareAlike 4.0 International License（CC BY-SA 4.0）授权。您可以自由分享（复制与再分发）及改编（混编、转换、在其基础上创作）本材料，可用于任何目的（包括商业用途） ，前提是给予适当的署名、提供许可证的链接、指明是否做了修改，并以相同许可证分发任何衍生作品。完整许可证文本：https://creativecommons.org/licenses/by-sa/4.0/。翻译声明。本中文版由 Jiancong Zhu（@Chasing1020）借助大语言模型辅助翻译完成，并经人工审校。翻译过程中保留原文的论述结构、技术内容与引用关系。本中文译本作为原作的衍生作品，同样依据 Creative Commons Attribution-ShareAlike 4.0 International License（CC BY-SA 4.0）发布；任何对本译本的再分发或改编必须保留对原作者 Haggai Roitman 与原作的署名，并以相同许可证授权。译本仓库与勘误：github.com/Chasing1020/agentic-ai-guide-zh；如发现翻译错误或表达不当，欢迎通过 GitHub Issue 或 chasing1020@gmail.com 反馈以便修订。关于作者Haggai Roitman 在 AI 研究与大规模生产系统的交叉点上耕耘了二十余年。他的工作贯通理论与实践——从发表基础性研究，到交付服务数百万用户的系统。他的研究兴趣涵盖信息检索、推荐系统、自然语言处理、大语言模型、面向 LLM 的强化学习以及智能体 AI。他已发表逾 100 篇同行评审论文，并拥有约 100 项专利。他在 Technion——以色列理工学院取得本科（Cum Laude 优等）与博士学位。当他不在思考梯度流与 KV 缓存时，可以在打碟机前找到他——在那里他混音 progressive trance与 deep house。

「预测未来的最好方式就是亲手创造它。」

—— Alan Kay

参考文献

  1. [1]A. Abid, A. Abdalla, A. Abid, D. Khan, A. Alfozan, and J. Zou (2019) Gradio: Hassle-Free Sharing and Testing of ML Models in the Wild. 外部链接：链接
  2. [2]S. Abnar and W. Zuidema (2020) Quantifying Attention Flow in Transformers. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL), 外部链接：链接
  3. [3]A. Aghajanyan, S. Gupta, and L. Zettlemoyer (2021) Intrinsic Dimensionality Explains the Effectiveness of Language Model Fine-Tuning. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL), 外部链接：链接
  4. [4]M. AI (2025) The Llama 4 Herd: The Beginning of a New Era of Natively Multimodal AI. Meta AI Blog. 外部链接：链接
  5. [5]M. AI (2024) Mistral Large 2. Mistral AI Blog. 外部链接：链接
  6. [6]D. Amodei, C. Olah, J. Steinhardt, P. Christiano, J. Schulman, and D. Mané (2016) Concrete Problems in AI Safety. arXiv Preprint arXiv:1606.06565. 外部链接：链接
  7. [7]Y. An, L. Shen, Y. Xu, and X. Liu (2025) DPPO: Direct Divergence-Based Policy Optimization for Language Model Alignment. arXiv Preprint arXiv:2503.14532. 外部链接：链接
  8. [8]M. Andrychowicz, F. Wolski, A. Ray, et al. (2017) Hindsight Experience Replay. In Advances in Neural Information Processing Systems (NeurIPS),
  9. [9]Anthropic (2024) Building Effective Agents. 外部链接：链接
  10. [10]Anthropic (2024) Model Context Protocol. 外部链接：链接
  11. [11]Anthropic (2024) The Claude 3 Model Family: Opus, Sonnet, Haiku. Anthropic Technical Report. 外部链接：链接
  12. [12]Anthropic (2026) Natural Language Autoencoders: Interpreting Neural Networks with Natural Language Descriptions. Anthropic Research Blog. 外部链接：链接
  13. [13]A. Asai, Z. Wu, Y. Wang, A. Sil, and H. Hajishirzi (2023) Self-RAG: Learning to Retrieve, Generate, and Critique Through Self-Reflection. arXiv Preprint arXiv:2310.11511. 外部链接：链接
  14. [14]M. G. Azar, M. Rowland, B. Piot, et al. (2024) A General Theoretical Paradigm to Understand Learning from Human Feedback. arXiv Preprint arXiv:2310.12036. 外部链接：链接
  15. [15]J. L. Ba, J. R. Kiros, and G. E. Hinton (2016) Layer Normalization. arXiv Preprint arXiv:1607.06450. 外部链接：链接
  16. [16]Y. Bai, A. Jones, K. Ndousse, et al. (2022) Constitutional AI: Harmlessness from AI Feedback. arXiv Preprint arXiv:2212.08073. 外部链接：链接
  17. [17]Z. Bai Y. Bao et al. (2025) Kimi K2: Open Agentic Intelligence. arXiv preprint arXiv:2507.20534. 外部链接：链接
  18. [18]P. Bajaj, D. Campos, N. Craswell, et al. (2016) MS MARCO: A Human Generated MAchine Reading COmprehension Dataset. arXiv Preprint arXiv:1611.09268.
  19. [19]S. Banerjee and A. Lavie (2005) METEOR: An Automatic Metric for MT Evaluation with Improved Correlation with Human Judgments. In Proceedings of the ACL Workshop on Intrinsic and Extrinsic Evaluation Measures for Machine Translation and/or Summarization, 外部链接：链接
  20. [20]O. Barkan, E. Hauon, A. Caciularu, I. Dagan, and N. Koenigstein (2021) Grad-SAM: Explaining Transformers via Gradient Self-Attention Maps. In Proceedings of the 30th ACM International Conference on Information and Knowledge Management (CIKM), 外部链接：链接
  21. [21]A. G. Barto and S. Mahadevan (2003) Recent Advances in Hierarchical Reinforcement Learning. Discrete Event Dynamic Systems13, pp. 341–379.
  22. [22]F. L. Bellifemine, G. Caire, and D. Greenwood (2007) JADE: Developing Multi-Agent Systems with JADE.
  23. [23]I. Beltagy, M. E. Peters, and A. Cohan (2020) Longformer: The Long-Document Transformer. arXiv Preprint arXiv:2004.05150. 外部链接：链接
  24. [24]T. Berners-Lee, J. Hendler, and O. Lassila (2001) The Semantic Web. Scientific American.
  25. [25]M. Besta, N. Blach, A. Kubicek, et al. (2024) Graph of Thoughts: Solving Elaborate Problems with Large Language Models. In Proceedings of the AAAI Conference on Artificial Intelligence,
  26. [26]L. Beurer-Kellner, M. Fischer, and M. Vechev (2023) Prompting Is Programming: A Query Language for Large Language Models. In Proceedings of PLDI,
  27. [27]R. Bonatti, D. Zhao, F. Bonacci, et al. (2024) Windows Agent Arena: Evaluating Multi-Modal OS Agents at Scale. arXiv Preprint arXiv:2409.08264. 外部链接：链接
  28. [28]E. d. Bono (1985) Six Thinking Hats.
  29. [29]R. A. Bradley and M. E. Terry (1952) Rank Analysis of Incomplete Block Designs: I. The Method of Paired Comparisons. Biometrika. 外部链接：链接
  30. [30]T. Bricken, A. Templeton, J. Batson, et al. (2023) Towards Monosemanticity: Decomposing Language Models with Dictionary Learning. Transformer Circuits Thread. 外部链接：链接
  31. [31]G. Brockman, V. Cheung, L. Pettersson, et al. (2016) OpenAI Gym.
  32. [32]T. Brown, B. Mann, N. Ryder, et al. (2020) Language Models Are Few-Shot Learners. NeurIPS.
  33. [33]C. Burges, T. Shaked, E. Renshaw, et al. (2005) Learning to Rank Using Gradient Descent. In Proceedings of the 22nd International Conference on Machine Learning (ICML),
  34. [34]C. J. C. Burges, R. Ragno, and Q. V. Le (2006) Learning to Rank with Nonsmooth Cost Functions. In Advances in Neural Information Processing Systems (NeurIPS),
  35. [35]T. Cai, Y. Li, Z. Geng, et al. (2024) Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads. In Proceedings of the 41st International Conference on Machine Learning (ICML), 外部链接：链接
  36. [36]B. Campbell, J. Bradley, N. Sakimura, and T. Lodderstedt (2020) OAuth 2.0 Mutual-TLS Client Authentication and Certificate-Bound Access Tokens. 外部链接：链接
  37. [37]M. Campbell, A. J. H. Jr., and F. Hsu (2002) Deep Blue. Artificial Intelligence.
  38. [38]Z. Cao, T. Qin, T. Liu, M. Tsai, and H. Li (2007) Learning to Rank: From Pairwise Approach to Listwise Approach. In Proceedings of the 24th International Conference on Machine Learning (ICML),
  39. [39]N. Carlini, F. Tramer, E. Wallace, et al. (2021) Extracting Training Data from Large Language Models. USENIX Security Symposium. 外部链接：链接
  40. [40]Chainlit (2024) Chainlit: Build Production-Ready Conversational AI Applications. 外部链接：链接
  41. [41]H. Chase (2022) LangChain. 外部链接：链接
  42. [42]K. Chawla, J. Ramirez, R. Clever, G. Lucas, J. May, and J. Gratch (2021) CaSiNo: A Corpus of Campsite Negotiation Dialogues for Automatic Negotiation Systems. In Proceedings of NAACL,
  43. [43]H. Chen, G. Zheng, Y. Kim, and Y. Chen (2024) Noise Contrastive Alignment of Language Models with Explicit Rewards. arXiv Preprint arXiv:2402.05369. 外部链接：链接
  44. [44]J. Chen, S. Luo, M. Zhang, Z. Liu, Y. Xiao, and D. Han (2022) Hybrid Retrieval for Open-Domain Question Answering. arXiv Preprint arXiv:2210.06029. 外部链接：链接
  45. [45]J. Chen, S. Xiao, P. Zhang, K. Luo, D. Lian, and Z. Liu (2024) BGE M3-Embedding: Multi-Lingual, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation. arXiv Preprint arXiv:2402.03216.
  46. [46]L. Chen, M. Zaharia, and J. Zou (2023) FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance. arXiv Preprint arXiv:2305.05176. 外部链接：链接
  47. [47]M. Chen, J. Tworek, H. Jun, et al. (2021) Evaluating Large Language Models Trained on Code. arXiv Preprint arXiv:2107.03374. 外部链接：链接
  48. [48]S. Chen, S. Wong, L. Chen, and Y. Tian (2023) Extending Context Window of Large Language Models via Positional Interpolation. arXiv Preprint arXiv:2306.15595. 外部链接：链接
  49. [49]T. Chen, B. Xu, C. Zhang, and C. Guestrin (2016) Training Deep Nets with Sublinear Memory Cost. arXiv Preprint arXiv:1604.06174.
  50. [50]Y. Chen et al. (2026) MOSAIC: Safe Agentic Training via Plan-Check-Act-Refuse. Note: arXiv preprint
  51. [51]Z. Chen, Y. Deng, R. Zhang, H. Sun, and W. Chen (2025) GSPO: Sequence-Level Policy Optimization for Language Model Alignment. arXiv Preprint arXiv:2502.12459. 外部链接：链接
  52. [52]B. Cherny (2026) Loops Are the Next Step from Agents. Note: Talk at Meta @Scale ConferenceHead of Claude Code, Anthropic
  53. [53]P. Chhikara, D. Khant, S. Aryan, T. Singh, and D. Yadav (2025) Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory. arXiv Preprint arXiv:2504.19413. 外部链接：链接
  54. [54]S. Choi, H. Park, D. Moon, K. Kim, and E. Choi (2025) GOPO: Group Ordinal Policy Optimization for LLM Alignment with Non-Verifiable Rewards. arXiv Preprint arXiv:2505.12948. 外部链接：链接
  55. [55]A. Chowdhery, S. Narang, J. Devlin, et al. (2022) PaLM: Scaling Language Modeling with Pathways. arXiv Preprint arXiv:2204.02311.
  56. [56]S. R. Chowdhury, A. Chakraborty, S. Natarajan, A. Agarwal, and D. Sontag (2024) Provably Robust DPO: Aligning Language Models with Noisy Feedback. arXiv Preprint arXiv:2403.00409. 外部链接：链接
  57. [57]P. F. Christiano, J. Leike, T. B. Brown, M. Martic, S. Legg, and D. Amodei (2017) Deep Reinforcement Learning from Human Preferences. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  58. [58]Y. Chuang, Y. Xie, H. Luo, Y. Kim, J. Glass, and P. He (2024) DoLA: Decoding by Contrasting Layers Improves Factuality in Large Language Models. In Proceedings of the 12th International Conference on Learning Representations (ICLR), 外部链接：链接
  59. [59]H. W. Chung, L. Hou, S. Longpre, et al. (2024) Scaling Instruction-Finetuned Language Models. Journal of Machine Learning Research. 外部链接：链接
  60. [60]J. Cohen (1960) A Coefficient of Agreement for Nominal Scales. Educational and Psychological Measurement.
  61. [61]W. W. W. Consortium (2022) Decentralized Identifiers (DIDs) v1.0. 外部链接：链接
  62. [62]G. V. Cormack, C. L. A. Clarke, and S. Buettcher (2009) Reciprocal Rank Fusion Outperforms Condorcet and Individual Rank Learning Methods. In Proceedings of the 32nd International ACM SIGIR Conference,
  63. [63]H. Cunningham, A. Ewart, L. Riggs, R. Huben, and L. Sharkey (2024) Sparse Autoencoders Find Highly Interpretable Features in Language Models. In Proceedings of the 12th International Conference on Learning Representations (ICLR), 外部链接：链接
  64. [64]T. Dao, D. Y. Fu, S. Ermon, A. Rudra, and C. Ré (2022) FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  65. [65]T. Dao (2024) FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning. In International Conference on Learning Representations (ICLR), 外部链接：链接
  66. [66]A. Das, T. Gerber, G. Gkioxari, S. Lee, D. Parikh, and D. Batra (2019) TarMAC: Targeted Multi-Agent Communication. In Proceedings of the 36th International Conference on Machine Learning (ICML), 外部链接：链接
  67. [67]Databricks (2024) DBRX: A New State-of-the-Art Open LLM. Databricks Blog. 外部链接：链接
  68. [68]G. DeepMind (2024) AlphaProof and AlphaGeometry 2: Solving Olympiad Geometry Without Human Demonstrations. 外部链接：链接
  69. [69]D. G. DeepSeek-AI D. Yang et al. (2025) DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning. arXiv Preprint arXiv:2501.12948. 外部链接：链接
  70. [70]DeepSeek-AI (2024) DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model. arXiv Preprint arXiv:2405.04434. 外部链接：链接
  71. [71]DeepSeek-AI (2024) DeepSeek-V3 Technical Report. arXiv Preprint arXiv:2412.19437. 外部链接：链接
  72. [72]DeepSeek-AI (2026) DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence. arXiv preprint arXiv:2606.19348. 外部链接：链接
  73. [73]X. Deng, Y. Gu, B. Zheng, et al. (2023) Mind2Web: Towards a Generalist Agent for the Web. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  74. [74]M. Dennis, N. Jaques, E. Vinitsky, et al. (2020) Emergent Complexity and Zero-Shot Transfer via Unsupervised Environment Design. In Advances in Neural Information Processing Systems (NeurIPS),
  75. [75]T. Dettmers, A. Pagnoni, A. Holtzman, and L. Zettlemoyer (2023) QLoRA: Efficient Finetuning of Quantized Language Models. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  76. [76]J. Devlin, M. Chang, K. Lee, and K. Toutanova (2019) BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of NAACL-HLT, 外部链接：链接
  77. [77]Y. Dong, C. Moon, Y. Wang, et al. (2024) XGrammar: Flexible and Efficient Structured Generation Engine for Large Language Models. arXiv Preprint arXiv:2411.15100.
  78. [78]Y. Du, S. Li, A. Torralba, J. B. Tenenbaum, and I. Mordatch (2023) Improving Factuality and Reasoning in Language Models Through Multiagent Debate. In Proceedings of the 41st International Conference on Machine Learning (ICML), 外部链接：链接
  79. [79]Y. Duan, J. Schulman, X. Chen, P. L. Bartlett, I. Sutskever, and P. Abbeel (2016) RL2: Fast Reinforcement Learning via Slow Reinforcement Learning. arXiv Preprint arXiv:1611.02779.
  80. [80]E. H. Durfee, V. R. Lesser, and D. D. Corkill (1989) Trends in Cooperative Distributed Problem Solving. IEEE Transactions on Knowledge and Data Engineering. 外部链接：链接
  81. [81]H. Ebbinghaus (1885) Über Das Gedächtnis: Untersuchungen Zur Experimentellen Psychologie.
  82. [82]D. Edge, H. Trinh, N. Cheng, et al. (2024) From Local to Global: A Graph RAG Approach to Query-Focused Summarization. arXiv Preprint arXiv:2404.16130. 外部链接：链接
  83. [83]V. Egiazarian, A. Panferov, D. Kuznedelev, E. Frantar, A. Babber, and D. Alistarh (2024) Extreme Compression of Large Language Models via Additive Quantization. arXiv Preprint arXiv:2401.06118. 外部链接：链接
  84. [84]A. E. Elo (1978) The Rating of Chess Players, Past and Present.
  85. [85]S. Es, J. James, L. Espinosa-Anke, and S. Schockaert (2023) RAGAs: Automated Evaluation of Retrieval Augmented Generation. arXiv Preprint arXiv:2309.15217. 外部链接：链接
  86. [86]K. Ethayarajh, W. Xu, N. Muennighoff, D. Jurafsky, and D. Kiela (2024) KTO: Model Alignment as Prospect Theoretic Optimization. arXiv Preprint arXiv:2402.01306. 外部链接：链接
  87. [87]K. Ethayarajh (2019) How Contextual Are Contextualized Word Representations? Comparing the Geometry of BERT, ELMo, and GPT-2 Embeddings. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing (EMNLP), 外部链接：链接
  88. [88]H. Face (2025) OpenEnv: An Interface Library for RL Post-Training with Environments. 外部链接：链接
  89. [89]W. Fedus, B. Zoph, and N. Shazeer (2022) Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity. Journal of Machine Learning Research.
  90. [90]D. Ferrucci, E. Brown, J. Chu-Carroll, et al. (2010) Building Watson: An Overview of the DeepQA Project. AI Magazine.
  91. [91]J. R. Firth (1957) A Synopsis of Linguistic Theory, 1930–1955. Studies in Linguistic Analysis.
  92. [92]J. L. Fleiss (1971) Measuring Nominal Scale Agreement Among Many Raters. Psychological Bulletin.
  93. [93]F. for Intelligent Physical Agents (2002) FIPA ACL Message Structure Specification. 外部链接：链接
  94. [94]T. Formal, C. Lassance, B. Piwowarski, and S. Clinchant (2021) SPLADE v2: Sparse Lexical and Expansion Model for Information Retrieval. arXiv Preprint arXiv:2109.10086.
  95. [95]T. Formal, B. Piwowarski, and S. Clinchant (2021) SPLADE: Sparse Lexical and Expansion Model for First Stage Ranking. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval,
  96. [96]T. Forte (2022) Building a Second Brain: A Proven Method to Organize Your Digital Life and Unlock Your Creative Potential.
  97. [97]N. Francis, A. Green, P. Guagliardo, et al. (2018) Cypher: An Evolving Query Language for Property Graphs. In Proceedings of the 2018 International Conference on Management of Data (SIGMOD),
  98. [98]E. Frantar and D. Alistarh (2023) SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot. In Proceedings of the 40th International Conference on Machine Learning (ICML), 外部链接：链接
  99. [99]E. Frantar, S. Ashkboos, T. Hoefler, and D. Alistarh (2023) GPTQ: Accurate Post-Training Quantization for Generative Pre-Trained Transformers. arXiv Preprint arXiv:2210.17323. 外部链接：链接
  100. [100]Y. Fu, R. Panda, X. Niu, et al. (2024) Data Engineering for Scaling Language Models to 128K Context. arXiv Preprint arXiv:2402.10171. 外部链接：链接
  101. [101]Y. Fu, P. Bailis, I. Stoica, and H. Zhang (2024) Break the Sequential Dependency of LLM Inference Using Lookahead Decoding. arXiv Preprint arXiv:2402.02057. 外部链接：链接
  102. [102]P. Furgale, S. Klingler, J. Nolan, M. Staats, G. Di Lorenzo, E. Martinez Abad, C. Schueler, R. Dinu, A. Devoto, P. Berard, G. Kaplun, E. Sarafian, R. Roveri, L. Derczynski, and R. Silveira Cabral (2026) NVIDIA-labs OO Agents: Native Python Object-Oriented Agents. Technical report Note: arXiv preprint arXiv:2607.20709 外部链接：链接
  103. [103]A. Gal, A. Anaby-Tavor, A. Trombetta, and D. Montesi (2005) A Framework for Modeling and Evaluating Automatic Semantic Reconciliation. In Proceedings of the 31st International Conference on Very Large Data Bases (VLDB), 外部链接：链接
  104. [104]I. O. Gallegos, R. A. Rossi, J. Barber, S. Tong, et al. (2024) Bias and Fairness in Large Language Models: A Survey. Computational Linguistics. 外部链接：链接
  105. [105]L. Gao, J. Schulman, and J. Hilton (2023) Scaling Laws for Reward Model Overoptimization. In Proceedings of the 40th International Conference on Machine Learning (ICML),
  106. [106]L. Gao, X. Ma, J. Lin, and J. Callan (2023) Precise Zero-Shot Dense Retrieval Without Relevance Labels. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL), 外部链接：链接
  107. [107]J. García and F. Fernández (2015) A Comprehensive Survey on Safe Reinforcement Learning. Journal of Machine Learning Research.
  108. [108]G. Gemini Team (2024) Gemini 1.5: Unlocking Multimodal Understanding Across Millions of Tokens of Context. arXiv Preprint arXiv:2403.05530. 外部链接：链接
  109. [109]G. Gerganov (2023) GGUF: GPT-Generated Unified Format (llama.cpp). 外部链接：链接
  110. [110]M. Geva, R. Schuster, J. Berant, and O. Levy (2021) Transformer Feed-Forward Layers Are Key-Value Memories. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP),
  111. [111]F. Gloeckle, B. Y. Idrissi, B. Rozière, D. Lopez-Paz, and G. Synnaeve (2024) Better & Faster Large Language Models via Multi-Token Prediction. In Proceedings of the 41st International Conference on Machine Learning (ICML), 外部链接：链接
  112. [112]C. A. E. Goodhart (1984) Problems of Monetary Management: The U.K. Experience. Monetary Theory and Practice.
  113. [113]Google DeepMind (2026) Decoupled DiLoCo: Distributed Training Across Geographic Regions. Note: arXiv preprint
  114. [114]Google Research (2026) Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMs. Note: COLM 2026
  115. [115]Google (2025) Agent2Agent (A2A) Protocol. 外部链接：链接
  116. [116]A. Gorbatenko (2024) Online DPO with Synchronised Reference Model Updates. TRL Documentation.
  117. [117]P. Grassé (1959) La Reconstruction Du Nid Et Les Coordinations Interindividuelles Chez Bellicositermes Natalensis Et Cubitermes Sp. La Théorie de La Stigmergie: Essai d’interprétation Du Comportement Des Termites Constructeurs. Insectes Sociaux.
  118. [118]A. Grattafiori, A. Dubey, A. Jauhri, et al. (2024) The Llama 3 Herd of Models. arXiv Preprint arXiv:2407.21783. 外部链接：链接
  119. [119]A. Graves, G. Wayne, M. Reynolds, et al. (2016) Hybrid Computing Using a Neural Network with Dynamic External Memory. Nature.
  120. [120]K. Greshake, S. Abdelnabi, S. Mishra, C. Endres, T. Holz, and M. Fritz (2023) Not What You’ve Signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection. In Proceedings of the 16th ACM Workshop on Artificial Intelligence and Security (AISec),
  121. [121]J. W. Group (2010) JSON-RPC 2.0 Specification. 外部链接：链接
  122. [122]A. Gu and T. Dao (2023) Mamba: Linear-Time Sequence Modeling with Selective State Spaces. arXiv Preprint arXiv:2312.00752. 外部链接：链接
  123. [123]A. Gu and T. Dao (2024) Mamba: Linear-Time Sequence Modeling with Selective State Spaces. In Proceedings of the 41st International Conference on Machine Learning (ICML), 外部链接：链接
  124. [124]C. Guo, G. Pleiss, Y. Sun, and K. Q. Weinberger (2017) On Calibration of Modern Neural Networks. In International Conference on Machine Learning (ICML),
  125. [125]M. Guo, J. Ainslie, D. Uthus, et al. (2022) LongT5: Efficient Text-to-Text Transformer for Long Sequences. Findings of the Association for Computational Linguistics: NAACL 2022. 外部链接：链接
  126. [126]R. Guo, P. Sun, E. Lindgren, et al. (2020) Accelerating Large-Scale Inference with Anisotropic Vector Quantization. In International Conference on Machine Learning (ICML),
  127. [127]S. Guo, B. Zhang, T. Liu, et al. (2024) Direct Language Model Alignment from Online AI Feedback. arXiv Preprint arXiv:2402.04792. 外部链接：链接
  128. [128]K. Guu, K. Lee, Z. Tung, P. Pasupat, and M. Chang (2020) REALM: Retrieval-Augmented Language Model Pre-Training. In Proceedings of the 37th International Conference on Machine Learning (ICML), 外部链接：链接
  129. [129]T. Haarnoja, A. Zhou, P. Abbeel, and S. Levine (2018) Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor. In Proceedings of the 35th International Conference on Machine Learning (ICML), 外部链接：链接
  130. [130]D. Hafner, T. Lillicrap, J. Ba, and M. Norouzi (2020) Dream to Control: Learning Behaviors by Latent Imagination. In Proceedings of the 8th International Conference on Learning Representations (ICLR), 外部链接：链接
  131. [131]C. Han, M. Li, and W. Chen (2025) SAPO: Soft Adaptive Policy Optimization for Efficient LLM Alignment. arXiv Preprint arXiv:2503.01739. 外部链接：链接
  132. [132]D. Han and M. Han (2024) Unsloth: Efficient LLM Fine-Tuning. 外部链接：链接
  133. [133]D. Hardt (2012) The OAuth 2.0 Authorization Framework. 外部链接：链接
  134. [134]S. Harris and A. Seaborne (2013) SPARQL 1.1 Query Language. 外部链接：链接
  135. [135]B. Hayes-Roth (1985) A Blackboard Architecture for Control. Artificial Intelligence. 外部链接：链接
  136. [136]S. Hayou, N. Ghosh, and B. Yu (2024) LoRA+: Efficient Low Rank Adaptation of Large Models. arXiv Preprint arXiv:2402.12354. 外部链接：链接
  137. [137]D. Hendrycks, C. Burns, S. Basart, et al. (2021) Measuring Massive Multitask Language Understanding. In International Conference on Learning Representations (ICLR),
  138. [138]R. Herbrich, T. Minka, and T. Graepel (2006) TrueSkill™: A Bayesian Skill Rating System. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  139. [139]G. Hinton, O. Vinyals, and J. Dean (2015) Distilling the Knowledge in a Neural Network. arXiv Preprint arXiv:1503.02531.
  140. [140]J. Hoffmann, S. Borgeaud, A. Mensch, et al. (2022) Training Compute-Optimal Large Language Models. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  141. [141]J. Hong, N. Lee, and J. Thorne (2024) ORPO: Monolithic Preference Optimization Without Reference Model. arXiv Preprint arXiv:2403.07691. 外部链接：链接
  142. [142]S. Hong, M. Zhuge, J. Chen, et al. (2024) MetaGPT: Meta Programming for a Multi-Agent Collaborative Framework.
  143. [143]A. Hosseini, X. Yuan, P. Poupart, A. Trischler, and Y. Bengio (2024) V-STaR: Training Verifiers for Self-Taught Reasoners. arXiv Preprint arXiv:2402.06457.
  144. [144]N. Houlsby, A. Giber, S. Jastrzebski, et al. (2019) Parameter-Efficient Transfer Learning for NLP. In International Conference on Machine Learning (ICML), 外部链接：链接
  145. [145]P. Hsu, Y. Dai, V. Kothapalli, et al. (2024) Liger-Kernel: Efficient Triton Kernels for LLM Training. arXiv Preprint arXiv:2410.10989. 外部链接：链接
  146. [146]E. J. Hu, Y. Shen, P. Wallis, et al. (2022) LoRA: Low-Rank Adaptation of Large Language Models. In International Conference on Learning Representations (ICLR), 外部链接：链接
  147. [147]E. J. Hu, Y. Shen, P. Wallis, et al. (2022) LoRA: Low-Rank Adaptation of Large Language Models. arXiv Preprint arXiv:2106.09685.
  148. [148]J. Hu, X. Tao, W. Zhu, S. Yang, J. Liu, and Z. Li (2024) OpenRLHF: An Easy-to-Use, Scalable and High-Performance RLHF Framework. arXiv Preprint arXiv:2405.11143.
  149. [149]Q. Hu S. Yang et al. (2025) Taming the Long-Tail: Efficient Reasoning RL Training with Adaptive Drafter. arXiv Preprint arXiv:2511.16665. 外部链接：链接
  150. [150]S. Hu, Y. Tu, X. Han, et al. (2024) MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies. arXiv Preprint arXiv:2404.06395. 外部链接：链接
  151. [151]Q. Huang, J. Vora, P. Liang, and J. Leskovec (2024) MLAgentBench: Evaluating Language Agents on Machine Learning Experimentation. In Proceedings of the 41st International Conference on Machine Learning (ICML), 外部链接：链接
  152. [152]Y. Huang, Y. Cheng, A. Bapna, et al. (2019) GPipe: Efficient Training of Giant Neural Networks Using Pipeline Parallelism. In Advances in Neural Information Processing Systems (NeurIPS),
  153. [153]E. Hubinger, C. Denison, J. Mu, et al. (2024) Sleeper Agents: Training Deceptive LLMs That Persist Through Safety Training. arXiv Preprint arXiv:2401.05566.
  154. [154]L. Inc (2024) LangGraph Studio: The First Agent IDE. 外部链接：链接
  155. [155]L. Inc (2024) LangGraph: Build Stateful Multi-Actor Applications with LLMs. 外部链接：链接
  156. [156]S. Inc (2024) Streamlit: The Fastest Way to Build and Share Data Apps. 外部链接：链接
  157. [157]G. Irving, P. Christiano, and D. Amodei (2018) AI Safety via Debate. arXiv Preprint arXiv:1805.00899. 外部链接：链接
  158. [158]M. Jaderberg, W. M. Czarnecki, I. Dunning, et al. (2019) Human-Level Performance in 3D Multiplayer Games with Population-Based Reinforcement Learning. Science. 外部链接：链接
  159. [159]N. Jain, P. Chiang, Y. Wen, et al. (2024) NEFTune: Noisy Embeddings Improve Instruction Finetuning. In International Conference on Learning Representations (ICLR), 外部链接：链接
  160. [160]S. Jain and B. C. Wallace (2019) Attention Is Not Explanation. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), 外部链接：链接
  161. [161]E. Jang, S. Gu, and B. Poole (2017) Categorical Reparameterization with Gumbel-Softmax. In International Conference on Learning Representations (ICLR),
  162. [162]S. Jeong, J. Baek, S. Cho, S. J. Hwang, and J. C. Park (2024) Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models Through Question Complexity. arXiv Preprint arXiv:2403.14403. 外部链接：链接
  163. [163]J. Ji, A. Fisch, J. Weston, and S. Sukhbaatar (2024) Towards Exact Optimization of Language Model Alignment. arXiv Preprint arXiv:2402.05369. 外部链接：链接
  164. [164]Z. Ji, N. Lee, R. Frieske, et al. (2023) Survey of Hallucination in Natural Language Generation. ACM Computing Surveys. 外部链接：链接
  165. [165]A. Q. Jiang, A. Sablayrolles, A. Mensch, et al. (2023) Mistral 7B. arXiv Preprint arXiv:2310.06825. 外部链接：链接
  166. [166]A. Q. Jiang, A. Sablayrolles, A. Roux, et al. (2024) Mixtral of Experts. arXiv Preprint arXiv:2401.04088. 外部链接：链接
  167. [167]M. Jiang, E. Grefenstette, and T. Rocktäschel (2021) Prioritized Level Replay. In Proceedings of the 38th International Conference on Machine Learning (ICML), 外部链接：链接
  168. [168]Y. Jiang et al. (2025) Reasoning Beyond Chain-of-Thought: A Latent Computational View. Note: arXiv preprint arXiv:2601.08058 外部链接：链接
  169. [169]C. E. Jimenez, J. Yang, A. Wettig, et al. (2024) SWE-bench: Can Language Models Resolve Real-World GitHub Issues?. In International Conference on Learning Representations (ICLR), 外部链接：链接
  170. [170]B. Jin H. Zeng et al. (2025) Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning. arXiv Preprint arXiv:2503.09516.
  171. [171]J. Johnson, M. Douze, and H. Jégou (2021) Billion-Scale Similarity Search with GPUs. IEEE Transactions on Big Data.
  172. [172]M. Jones, J. Bradley, and N. Sakimura (2015) JSON Web Token (JWT). 外部链接：链接
  173. [173]M. Joshi, E. Choi, D. Weld, and L. Zettlemoyer (2017) TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension. In Proceedings of ACL,
  174. [174]J. Jumper, R. Evans, A. Pritzel, et al. (2021) Highly Accurate Protein Structure Prediction with AlphaFold. Nature.
  175. [175]S. Kadavath, T. Conerly, A. Askell, et al. (2022) Language Models (Mostly) Know What They Know. arXiv Preprint arXiv:2207.05221. 外部链接：链接
  176. [176]D. Kalajdzievski (2023) A Rank Stabilization Scaling Factor for Fine-Tuning with LoRA. arXiv Preprint arXiv:2312.03732. 外部链接：链接
  177. [177]J. Kaplan, S. McCandlish, T. Henighan, et al. (2020) Scaling Laws for Neural Language Models. arXiv Preprint arXiv:2001.08361. 外部链接：链接
  178. [178]A. Karpathy (2025) Context Engineering for AI Agents. Note: Social media postJune 2025
  179. [179]A. Karpathy (2026) AutoResearch: Automated ML Experimentation via Agent Loops. Note: Open-source project 外部链接：链接
  180. [180]V. Karpukhin, B. Oğuz, S. Min, et al. (2020) Dense Passage Retrieval for Open-Domain Question Answering. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), 外部链接：链接
  181. [181]J. Ke, Z. Wen, W. Li, C. He, and L. Zhang (2026) Respecting Self-Uncertainty in On-Policy Self-Distillation for Efficient LLM Reasoning. arXiv preprint arXiv:2605.13255. 外部链接：链接
  182. [182]O. Khattab, A. Singhvi, P. Maheshwari, et al. (2024) DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines. In Proceedings of the 12th International Conference on Learning Representations (ICLR), 外部链接：链接
  183. [183]O. Khattab and M. Zaharia (2020) ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT. In Proceedings of the 43rd International ACM SIGIR Conference, 外部链接：链接
  184. [184]S. Kim S. J. Jang et al. (2023) Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes. Findings of the ACL. 外部链接：链接
  185. [185]D. P. Kingma and J. Ba (2015) Adam: A Method for Stochastic Optimization. In International Conference on Learning Representations (ICLR), 外部链接：链接
  186. [186]J. Kirkpatrick, R. Pascanu, N. Rabinowitz, et al. (2017) Overcoming Catastrophic Forgetting in Neural Networks. In Proceedings of the National Academy of Sciences,
  187. [187]L. Kocsis and C. Szepesvári (2006) Bandit Based Monte-Carlo Planning. In European Conference on Machine Learning (ECML),
  188. [188]J. Y. Koh, R. Lo, L. Jang, et al. (2024) VisualWebArena: Evaluating Multimodal Agents on Realistic Visual Web Tasks. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL), 外部链接：链接
  189. [189]T. Kojima, S. S. Gu, M. Reid, Y. Matsuo, and Y. Iwasawa (2022) Large Language Models Are Zero-Shot Reasoners. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  190. [190]D. J. Kopiczko, T. Blankevoort, and M. Nagel (2024) VeRA: Vector-Based Random Matrix Adaptation. arXiv Preprint arXiv:2310.11454. 外部链接：链接
  191. [191]V. A. Korthikanti, J. Casper, S. Lym, et al. (2023) Reducing Activation Recomputation in Large Transformer Models. In Proceedings of Machine Learning and Systems (MLSys),
  192. [192]A. Krizhevsky, I. Sutskever, and G. E. Hinton (2012) ImageNet Classification with Deep Convolutional Neural Networks. NeurIPS.
  193. [193]L. Kuhn, Y. Gal, and S. Farquhar (2023) Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation. In International Conference on Learning Representations (ICLR), 外部链接：链接
  194. [194]H. Küttler, N. Nardelli, A. H. Miller, et al. (2020) The NetHack Learning Environment. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  195. [195]T. Kwiatkowski, J. Palomaki, O. Redfield, et al. (2019) Natural Questions: A Benchmark for Question Answering Research. Transactions of the Association for Computational Linguistics.
  196. [196]W. Kwon, Z. Li, S. Zhuang, et al. (2023) Efficient Memory Management for Large Language Model Serving with PagedAttention. In Proceedings of the ACM SIGOPS 29th Symposium on Operating Systems Principles (SOSP), 外部链接：链接
  197. [197]T. Lacroix, G. Obozinski, and N. Usunier (2020) Tensor Decompositions for Temporal Knowledge Base Completion. In International Conference on Learning Representations (ICLR), 外部链接：链接
  198. [198]J. Lála, O. O’Donoghue, A. Shtedritski, S. Cox, S. G. Rodriques, and A. D. White (2023) PaperQA: Retrieval-Augmented Generative Agent for Scientific Research. arXiv Preprint arXiv:2312.07559. 外部链接：链接
  199. [199]N. Lambert, J. Morrison, V. Pyatkin, et al. (2024) Tülu 3: Pushing Frontiers in Open Language Model Post-Training. arXiv Preprint arXiv:2411.15124. 外部链接：链接
  200. [200]A. Lazaridou and M. Baroni (2020) Emergent Multi-Agent Communication in the Deep Learning Era. arXiv Preprint arXiv:2006.02419. 外部链接：链接
  201. [201]H. Le, Y. Wang, A. D. Gotmare, S. Savarese, and S. Hoi (2023) CodeRL: Mastering Code Generation Through Pretrained Models and Deep Reinforcement Learning. In Advances in Neural Information Processing Systems (NeurIPS),
  202. [202]T. Le Sellier De Chezelles, M. Gasse, A. Drouin, M. Caccia, et al. (2024) The BrowserGym Ecosystem for Web Agent Research. arXiv Preprint arXiv:2412.05467. 外部链接：链接
  203. [203]C. Lee, R. Roy, M. Xu, et al. (2024) NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models. arXiv Preprint arXiv:2405.17428.
  204. [204]K. Lee, D. Ippolito, A. Nystrom, et al. (2022) Deduplicating Training Data Makes Language Models Better. In Proceedings of the 60th Annual Meeting of the ACL, 外部链接：链接
  205. [205]B. Lester, R. Al-Rfou, and N. Constant (2021) The Power of Scale for Parameter-Efficient Prompt Tuning. In Proceedings of the 2021 Conference on EMNLP, 外部链接：链接
  206. [206]Y. Leviathan, M. Kalman, and Y. Matias (2023) Fast Inference from Transformers via Speculative Decoding. In Proceedings of the 40th International Conference on Machine Learning (ICML), 外部链接：链接
  207. [207]M. Lewis, Y. Liu, N. Goyal, et al. (2020) BART: Denoising Sequence-to-Sequence Pre-Training for Natural Language Generation, Translation, and Comprehension. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL), 外部链接：链接
  208. [208]M. Lewis, D. Yarats, Y. N. Dauphin, D. Parikh, and D. Batra (2017) Deal or No Deal? End-to-End Learning for Negotiation Dialogues. In Proceedings of EMNLP,
  209. [209]P. Lewis, E. Perez, A. Piktus, et al. (2020) Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  210. [210]S. Li, Y. Zhao, R. Varma, et al. (2020) PyTorch Distributed: Experiences on Accelerating Data Parallel Training. In Proceedings of the VLDB Endowment,
  211. [211]X. Li, A. Holtzman, D. Fried, et al. (2023) Contrastive Decoding: Open-Ended Text Generation as Optimization. ACL.
  212. [212]X. L. Li and P. Liang (2021) Prefix-Tuning: Optimizing Continuous Prompts for Generation. In Proceedings of the 59th Annual Meeting of the ACL, 外部链接：链接
  213. [213]Y. Li, F. Wei, C. Zhang, and H. Zhang (2024) EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty. In Proceedings of the 41st International Conference on Machine Learning (ICML), 外部链接：链接
  214. [214]Z. Li, X. Zhang, Y. Zhang, D. Long, P. Xie, and M. Zhang (2023) Towards General Text Embeddings with Multi-Stage Contrastive Learning. arXiv Preprint arXiv:2308.03281.
  215. [215]Z. Li et al. (2026) UniClawBench: A Bilingual Benchmark for Interactive Agent Evaluation. Note: arXiv preprint
  216. [216]H. Lightman, V. Kosaraju, Y. Burda, et al. (2023) Let’s Verify Step by Step. arXiv Preprint arXiv:2305.20050.
  217. [217]C. Lin (2004) ROUGE: A Package for Automatic Evaluation of Summaries. In Text Summarization Branches Out: Proceedings of the ACL-04 Workshop, 外部链接：链接
  218. [218]J. Lin, J. Tang, H. Tang, et al. (2024) AWQ: Activation-Aware Weight Quantization for LLM Compression and Acceleration. In Proceedings of Machine Learning and Systems (MLSys), 外部链接：链接
  219. [219]K. Lin, C. Snell, Y. Wang, et al. (2025) Sleep-Time Compute: Beyond Inference Scaling at Test-Time. arXiv Preprint arXiv:2504.13171. 外部链接：链接
  220. [220]L. Lin (1992) Self-Improving Reactive Agents Based on Reinforcement Learning, Planning and Teaching. Machine Learning.
  221. [221]X. Lin, A. Ghosh, B. K. H. Low, A. Shrivastava, and V. Mohan (2025) REFRAG: Rethinking RAG Based Decoding. arXiv Preprint arXiv:2509.01092.
  222. [222]H. Liu, M. Zaharia, and P. Abbeel (2023) Ring Attention with Blockwise Transformers for Near-Infinite Context. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  223. [223]H. Liu, D. Tam, M. Muqeeth, et al. (2022) Few-Shot Parameter-Efficient Fine-Tuning Is Better and Cheaper Than in-Context Learning. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  224. [224]J. Liu, D. Shen, Y. Zhang, B. Dolan, L. Carin, and W. Chen (2022) What Makes Good in-Context Examples for GPT-3?. In Proceedings of Deep Learning Inside Out (DeeLIO), ACL Workshop, 外部链接：链接
  225. [225]K. J. Liu, Y. Jin, Z. Zhou, et al. (2025) Muon: An Optimizer for Hidden Layers in Neural Networks. Note: arXiv preprint 外部链接：链接
  226. [226]N. F. Liu, K. Lin, J. Hewitt, et al. (2024) Lost in the Middle: How Language Models Use Long Contexts. Transactions of the Association for Computational Linguistics. 外部链接：链接
  227. [227]N. F. Liu, K. Lin, J. Hewitt, et al. (2024) Lost in the Middle: How Language Models Use Long Contexts. Transactions of the Association for Computational Linguistics. 外部链接：链接
  228. [228]Q. e. a. Liu (2025) Answer Convergence as a Signal for Early Stopping in Reasoning. In Proceedings of EMNLP, 外部链接：链接
  229. [229]S. Liu, C. Wang, H. Yin, et al. (2024) DoRA: Weight-Decomposed Low-Rank Adaptation. arXiv Preprint arXiv:2402.09353. 外部链接：链接
  230. [230]X. Liu, H. Yu, H. Zhang, et al. (2023) AgentBench: Evaluating LLMs as Agents. arXiv Preprint arXiv:2308.03688.
  231. [231]X. e. a. Liu (2025) Learning Like Humans: Advancing LLM Reasoning Capabilities via Adaptive Difficulty Curriculum Learning. arXiv Preprint arXiv:2505.08364. 外部链接：链接
  232. [232]Y. Liu, D. Iter, Y. Xu, S. Wang, R. Xu, and C. Zhu (2023) G-Eval: NLG Evaluation Using GPT-4 with Better Human Alignment. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP), 外部链接：链接
  233. [233]Y. Liu, L. Han, Y. Tan, et al. (2025) Understanding and Mitigating the Pretraining Distribution Bias in GRPO. arXiv Preprint arXiv:2505.07888. 外部链接：链接
  234. [234]Y. Liu, M. Ott, N. Goyal, et al. (2019) RoBERTa: A Robustly Optimized BERT Pretraining Approach. arXiv Preprint arXiv:1907.11692. 外部链接：链接
  235. [235]Z. Liu, B. Oguz, C. Zhao, et al. (2023) LLM-QAT: Data-Free Quantization Aware Training for Large Language Models. arXiv Preprint arXiv:2305.17888. 外部链接：链接
  236. [236]Z. Liu H. Dong et al. (2025) DLER: Doing Length Penalty Right—Incentivizing More Intelligence Per Token. arXiv Preprint arXiv:2510.15110. 外部链接：链接
  237. [237]Z. Liu et al. (2025) OctoThinker: Mid-Training Determines RL-Readiness of Base Models. Note: arXiv preprint
  238. [238]Z. Liu, J. Yuan, H. Jin, et al. (2024) KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache. In International Conference on Machine Learning (ICML), 外部链接：链接
  239. [239]I. Loshchilov and F. Hutter (2019) Decoupled Weight Decay Regularization. arXiv Preprint arXiv:1711.05101. 外部链接：链接
  240. [240]R. Lowe, Y. Wu, A. Tamar, J. Harb, P. Abbeel, and I. Mordatch (2017) Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  241. [241]C. Lu, C. Lu, R. T. Lange, J. Foerster, J. Clune, and D. Ha (2024) The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery. arXiv Preprint arXiv:2408.06292. 外部链接：链接
  242. [242]Y. Lu, M. Bartolo, A. Moore, S. Riedel, and P. Stenetorp (2022) Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (ACL), 外部链接：链接
  243. [243]Z. Luo, Z. Chen, Y. Jiang, et al. (2025) ScaleRL: Scaling Reinforcement Learning for LLM Reasoning. arXiv Preprint arXiv:2505.16356. 外部链接：链接
  244. [244]Z. Luo, J. Shi, T. Yu, and M. Chen (2025) VESPO: Variational Sequence-Level Soft Policy Optimization for LLM Alignment. arXiv Preprint arXiv:2505.07508. 外部链接：链接
  245. [245]T. Lütke (2025) Context Engineering. Note: Social media postCEO of Shopify, June 2025
  246. [246]A. Madaan, N. Tandon, P. Gupta, et al. (2023) Self-Refine: Iterative Refinement with Self-Feedback. In Advances in Neural Information Processing Systems (NeurIPS),
  247. [247]A. Maharana, D. Lee, S. Tuber, M. Ruber, F. Barbieri, and M. Bansal (2024) LoCoMo: Long-Context Conversation with Memory Operations. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP),
  248. [248]Yu. A. Malkov and D. A. Yashunin (2020) Efficient and Robust Approximate Nearest Neighbor Search Using Hierarchical Navigable Small World Graphs. IEEE Transactions on Pattern Analysis and Machine Intelligence.
  249. [249]P. Manakul, A. Liusie, and M. J. F. Gales (2023) SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models. In Proceedings of EMNLP, 外部链接：链接
  250. [250]S. McCandlish, J. Kaplan, D. Amodei, and O. D. Team (2018) An Empirical Model of Large-Batch Training. arXiv Preprint arXiv:1812.06162.
  251. [251]J. e. a. Mei (2025) APRIL: Active Partial Rollouts in Reinforcement Learning to Tame Long-Tail Generation. arXiv Preprint arXiv:2509.18521. 外部链接：链接
  252. [252]Y. Meng, M. Xia, and D. Chen (2024) SimPO: Simple Preference Optimization with a Reference-Free Reward. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  253. [253]Meta AI (2026) Remember When It Matters: Proactive Memory for Long-Horizon Agents. Note: arXiv preprint
  254. [254]G. Mialon, C. Fourrier, C. Swift, T. Wolf, Y. LeCun, and T. Scialom (2024) GAIA: A Benchmark for General AI Assistants. In International Conference on Learning Representations (ICLR), 外部链接：链接
  255. [255]P. Micikevicius, S. Narang, J. Alben, et al. (2018) Mixed Precision Training. In International Conference on Learning Representations (ICLR),
  256. [256]Microsoft (2023) Semantic Kernel: SDK for Integrating AI Models into Applications. 外部链接：链接
  257. [257]S. Min, X. Lyu, A. Holtzman, et al. (2022) Rethinking the Role of Demonstrations: What Makes in-Context Learning Work?. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP), 外部链接：链接
  258. [258]V. Mnih, A. P. Badia, M. Mirza, et al. (2016) Asynchronous Methods for Deep Reinforcement Learning. In Proceedings of the 33rd International Conference on Machine Learning (ICML), 外部链接：链接
  259. [259]V. Mnih, K. Kavukcuoglu, D. Silver, et al. (2015) Human-Level Control Through Deep Reinforcement Learning. Nature. 外部链接：链接
  260. [260]MosaicML (2023) MPT-7B: A New Standard for Open-Source, Commercially Usable LLMs. MosaicML Blog. 外部链接：链接
  261. [261]J. Moura (2023) CrewAI: Framework for Orchestrating Role-Playing Autonomous AI Agents. 外部链接：链接
  262. [262]R. Nakano, J. Hilton, S. Balaji, et al. (2021) WebGPT: Browser-Assisted Question-Answering with Human Feedback. arXiv Preprint arXiv:2112.09332. 外部链接：链接
  263. [263]D. Narayanan, A. Harlap, A. Phanishayee, et al. (2019) PipeDream: Generalized Pipeline Parallelism for DNN Training. In Proceedings of the 27th ACM Symposium on Operating Systems Principles (SOSP),
  264. [264]D. Narayanan, M. Shoeybi, J. Casper, et al. (2021) Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM. arXiv Preprint arXiv:2104.04473.
  265. [265]A. Y. Ng, D. Harada, and S. J. Russell (1999) Policy Invariance Under Reward Transformations: Theory and Application to Reward Shaping. In Proceedings of the 16th International Conference on Machine Learning (ICML),
  266. [266]M. Nguyen (2024) Min-p Sampling: A Simple Baseline for Better LLM Decoding. arXiv Preprint arXiv:2310.06022.
  267. [267]N. Nisan, T. Roughgarden, É. Tardos, and V. V. Vazirani (2007) Algorithmic Game Theory. 外部链接：链接
  268. [268]R. Nogueira and K. Cho (2019) Passage Re-Ranking with BERT. arXiv Preprint arXiv:1901.04085.
  269. [269]R. Nogueira, Z. Jiang, R. Pradeep, and J. Lin (2020) Document Ranking with a Pretrained Sequence-to-Sequence Model. Findings of EMNLP.
  270. [270]NVIDIA (2025) Dynamo: A Datacenter-Scale Distributed Inference Serving Framework. Note: Open-source project 外部链接：链接
  271. [271]NVIDIA (2025) NeMo Gym: Interactive RL Environments for LLM Agent Training. Note: Technical report 外部链接：链接
  272. [272]M. Nye, A. J. Andreassen, G. Gur-Ari, et al. (2021) Show Your Work: Scratchpads for Intermediate Computation with Language Models. arXiv Preprint arXiv:2112.00114. 外部链接：链接
  273. [273]C. Olsson, N. Elhage, N. Nanda, et al. (2022) In-Context Learning and Induction Heads. Transformer Circuits Thread. 外部链接：链接
  274. [274]OpenAI (2023) GPT-4 Technical Report. arXiv Preprint arXiv:2303.08774.
  275. [275]OpenAI (2024) Learning to Reason with LLMs. 外部链接：链接
  276. [276]OpenAI (2024) Swarm: An Educational Framework for Lightweight Multi-Agent Orchestration. 外部链接：链接
  277. [277]OpenAI (2025) OpenAI Agents SDK. 外部链接：链接
  278. [278]OpenAI (2025) OpenAI o3 and o4-mini System Card. 外部链接：链接
  279. [279]A. Osmani (2026) Loop Engineering. Note: Blog post 外部链接：链接
  280. [280]L. Ouyang, J. Wu, X. Jiang, et al. (2022) Training Language Models to Follow Instructions with Human Feedback. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  281. [281]C. Packer, V. Fang, S. G. Patil, K. Lin, S. Wooders, and J. E. Gonzalez (2023) MemGPT: Towards LLMs as Operating Systems. arXiv Preprint arXiv:2310.08560. 外部链接：链接
  282. [282]L. Pan, S. Tao, Y. Zhai, L. Zhang, Z. Liu, B. Ding, A. Liu, and L. Wen (2026) RLCSD: Reinforcement Learning with Contrastive On-Policy Self-Distillation. arXiv preprint arXiv:2606.11709. 外部链接：链接
  283. [283]K. Papineni, S. Roukos, T. Ward, and W. Zhu (2002) BLEU: A Method for Automatic Evaluation of Machine Translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics (ACL), 外部链接：链接
  284. [284]R. Parasuraman and V. Riley (1997) Humans and Automation: Use, Misuse, Disuse, Abuse. Human Factors. 外部链接：链接
  285. [285]J. S. Park, J. C. O’Brien, C. J. Cai, M. R. Morris, P. Liang, and M. S. Bernstein (2023) Generative Agents: Interactive Simulacra of Human Behavior. In Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology (UIST), 外部链接：链接
  286. [286]D. Pathak, P. Agrawal, A. A. Efros, and T. Darrell (2017) Curiosity-Driven Exploration by Self-Supervised Prediction. In Proceedings of the 34th International Conference on Machine Learning (ICML),
  287. [287]S. G. Patil, T. Zhang, X. Wang, and J. E. Gonzalez (2024) Gorilla: Large Language Model Connected with Massive APIs. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  288. [288]B. Peng, E. Alcaide, Q. Anthony, et al. (2023) RWKV: Reinventing RNNs for the Transformer Era. Findings of the Association for Computational Linguistics: EMNLP 2023. 外部链接：链接
  289. [289]B. Peng, J. Quesnelle, H. Fan, and E. Shao (2023) YaRN: Efficient Context Window Extension of Large Language Models. arXiv Preprint arXiv:2309.00071.
  290. [290]E. Perez, S. Huang, F. Song, et al. (2022) Red Teaming Language Models with Language Models. In Proceedings of EMNLP, 外部链接：链接
  291. [291]R. L. Plackett (1975) The Analysis of Permutations. Journal of the Royal Statistical Society: Series C (Applied Statistics).
  292. [292]O. Press, N. A. Smith, and M. Lewis (2022) Train Short, Test Long: Attention with Linear Biases Enables Input Length Generalization. ICLR.
  293. [293]T. Preston-Werner (2024) Semantic Versioning 2.0.0. 外部链接：链接
  294. [294]P. Putta, E. Mills, N. Garg, et al. (2024) Agent Q: Advanced Reasoning and Learning for Autonomous AI Agents. arXiv Preprint arXiv:2408.07199.
  295. [295]PyTorch Team (2026) Miles: A PyTorch-Native Stack for Large-Scale LLM RL Post-Training. Note: PyTorch Blog 外部链接：链接
  296. [296]P. Qi, X. Wan, G. Huang, and M. Lin (2023) Zero Bubble Pipeline Parallelism. arXiv Preprint arXiv:2401.10241.
  297. [297]Z. Qi, M. Wan, J. Cao, and M. Lin (2024) Mutual Reasoning Makes Smaller LLMs Stronger Problem-Solvers. arXiv Preprint arXiv:2408.06195.
  298. [298]Y. Qin, X. Li, H. Zou, et al. (2024) O1 Replication Journey: A Strategic Progress Report. arXiv Preprint arXiv:2410.18982. 外部链接：链接
  299. [299]Y. Qin, S. Liang, Y. Ye, et al. (2024) ToolLLM: Facilitating Large Language Models to Master 16000+ Real-World APIs. In Proceedings of the 12th International Conference on Learning Representations (ICLR), 外部链接：链接
  300. [300]A. H. Rackauckas (2024) RAG-Fusion: A New Take on Retrieval-Augmented Generation.
  301. [301]A. Radford, J. Wu, R. Child, D. Luen, D. Amodei, and I. Sutskever (2019) Language Models Are Unsupervised Multitask Learners. OpenAI Blog. 外部链接：链接
  302. [302]R. Rafailov, A. Sharma, E. Mitchell, C. D. Manning, S. Ermon, and C. Finn (2023) Direct Preference Optimization: Your Language Model Is Secretly a Reward Model. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  303. [303]C. Raffel, N. Shazeer, A. Roberts, et al. (2020) Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer. Journal of Machine Learning Research. 外部链接：链接
  304. [304]S. Rajbhandari, J. Rasley, O. Rber, and Y. He (2020) ZeRO: Memory Optimizations Toward Training Trillion Parameter Models. arXiv Preprint arXiv:1910.02054.
  305. [305]S. Rajbhandari, O. Ruwase, J. Rasley, S. Smith, and Y. He (2021) ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning. arXiv Preprint arXiv:2104.07857.
  306. [306]T. Rashid, M. Samvelyan, C. S. d. Witt, G. Farquhar, J. Foerster, and S. Whiteson (2018) QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning. In Proceedings of the 35th International Conference on Machine Learning (ICML), 外部链接：链接
  307. [307]E. Rescorla (2018) The Transport Layer Security (TLS) Protocol Version 1.3. 外部链接：链接
  308. [308]H. Robbins and S. Monro (1951) A Stochastic Approximation Method. The Annals of Mathematical Statistics.
  309. [309]S. Robertson and H. Zaragoza (2009) The Probabilistic Relevance Framework: BM25 and Beyond. Foundations and Trends in Information Retrieval.
  310. [310]D. E. Rumelhart, G. E. Hinton, and R. J. Williams (1986) Learning Representations by Back-Propagating Errors. Nature. 外部链接：链接
  311. [311]G. A. Rummery and M. Niranjan (1994) On-Line q-Learning Using Connectionist Systems.
  312. [312]V. Sanh, L. Debut, J. Chaumond, and T. Wolf (2019) DistilBERT, a Distilled Version of BERT: Smaller, Faster, Cheaper and Lighter. arXiv Preprint arXiv:1910.01108.
  313. [313]K. Santhanam, O. Khattab, J. Saad-Falcon, C. Potts, and M. Zaharia (2022) ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics (NAACL),
  314. [314]T. Schaul, J. Quan, I. Antonoglou, and D. Silver (2016) Prioritized Experience Replay. In Proceedings of the 4th International Conference on Learning Representations (ICLR), 外部链接：链接
  315. [315]T. Schick, J. Dwivedi-Yu, R. Dessì, et al. (2023) Toolformer: Language Models Can Teach Themselves to Use Tools. In Advances in Neural Information Processing Systems,
  316. [316]J. Schrittwieser, I. Antonoglou, T. Hubert, et al. (2020) Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model. Nature. 外部链接：链接
  317. [317]J. Schulman, S. Levine, P. Abbeel, M. Jordan, and P. Moritz (2015) Trust Region Policy Optimization. In Proceedings of the 32nd International Conference on Machine Learning (ICML), 外部链接：链接
  318. [318]J. Schulman, P. Moritz, S. Levine, M. Jordan, and P. Abbeel (2016) High-Dimensional Continuous Control Using Generalized Advantage Estimation. In Proceedings of the 4th International Conference on Learning Representations (ICLR), 外部链接：链接
  319. [319]J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov (2017) Proximal Policy Optimization Algorithms. arXiv Preprint arXiv:1707.06347. 外部链接：链接
  320. [320]R. Sennrich, B. Haddow, and A. Birch (2016) Neural Machine Translation of Rare Words with Subword Units. In Proceedings of the 54th Annual Meeting of the ACL, 外部链接：链接
  321. [321]A. Sergeev and M. D. Balso (2018) Horovod: Fast and Easy Distributed Deep Learning in TensorFlow. arXiv Preprint arXiv:1802.05799.
  322. [322]J. Shah, G. Bikshandi, Y. Zhang, V. Thakkar, P. Ramani, and T. Dao (2024) FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-Precision. arXiv Preprint arXiv:2407.08691. 外部链接：链接
  323. [323]Z. Shao, P. Wang, Q. Zhu, et al. (2024) DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models. arXiv Preprint arXiv:2402.03300. 外部链接：链接
  324. [324]L. S. Shapley (1953) Stochastic Games. In Proceedings of the National Academy of Sciences, 外部链接：链接
  325. [325]N. Shazeer, A. Mirhoseini, K. Maziarz, et al. (2017) Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer. In International Conference on Learning Representations (ICLR), 外部链接：链接
  326. [326]N. Shinn, F. Cassano, A. Gopinath, K. Narasimhan, and S. Yao (2023) Reflexion: Language Agents with Verbal Reinforcement Learning. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  327. [327]M. Shoeybi, M. Patwary, R. Puri, P. LeGresley, J. Casper, and B. Catanzaro (2019) Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism. arXiv Preprint arXiv:1909.08053.
  328. [328]Y. Shoham and K. Leyton-Brown (2008) Multiagent Systems: Algorithmic, Game-Theoretic, and Logical Foundations. 外部链接：链接
  329. [329]M. Shridhar, X. Yuan, M. Côté, Y. Bisk, A. Trischler, and M. Hausknecht (2021) ALFWorld: Aligning Text and Embodied Environments for Interactive Learning. In International Conference on Learning Representations (ICLR),
  330. [330]Significant Gravitas (2023) AutoGPT: An Autonomous GPT-4 Experiment. Note: Open-source project 外部链接：链接
  331. [331]D. Silver, A. Huang, C. J. Maddison, et al. (2016) Mastering the Game of Go with Deep Neural Networks and Tree Search. Nature. 外部链接：链接
  332. [332]D. Silver, J. Schrittwieser, K. Simonyan, et al. (2017) Mastering the Game of Go Without Human Knowledge. Nature. 外部链接：链接
  333. [333]J. Skalse, N. Howe, D. Krasheninnikov, and D. Krueger (2022) Defining and Characterizing Reward Hacking. In Advances in Neural Information Processing Systems, 外部链接：链接
  334. [334]R. G. Smith (1980) The Contract Net Protocol: High-Level Communication and Control in a Distributed Problem Solver. IEEE Transactions on Computers. 外部链接：链接
  335. [335]C. Snell, J. Lee, K. Xu, and A. Kumar (2024) Scaling LLM Test-Time Compute Optimally Can Be More Effective Than Scaling Model Parameters. arXiv Preprint arXiv:2408.03314.
  336. [336]K. Sparck Jones (1972) A Statistical Interpretation of Term Specificity and Its Application in Retrieval. Journal of Documentation.
  337. [337]L. R. Squire (1992) Declarative and Nondeclarative Memory: Multiple Brain Systems Supporting Learning and Memory. Journal of Cognitive Neuroscience. 外部链接：链接
  338. [338]P. Steinberger (2026) OpenClaw: An Always-On Agent Orchestrator. Note: Open-source project 外部链接：链接
  339. [339]N. Stiennon, L. Ouyang, J. Wu, et al. (2020) Learning to Summarize from Human Feedback. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  340. [340]J. Su, M. Ahmed, Y. Lu, S. Pan, W. Bo, and Y. Liu (2024) RoFormer: Enhanced Transformer with Rotary Position Embedding. Neurocomputing.
  341. [341]J. Su, J. Cao, W. Liu, and Y. Ou (2021) Whitening Sentence Representations for Better Semantics and Faster Retrieval. arXiv Preprint arXiv:2103.15316. 外部链接：链接
  342. [342]S. Sukhbaatar, A. Szlam, and R. Fergus (2016) Learning Multiagent Communication with Backpropagation. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  343. [343]S. Sukhbaatar, A. Szlam, J. Weston, and R. Fergus (2015) End-to-End Memory Networks. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  344. [344]T. R. Sumers, S. Yao, K. Narasimhan, and T. L. Griffiths (2024) Cognitive Architectures for Language Agents. Transactions on Machine Learning Research (TMLR). 外部链接：链接
  345. [345]M. Sun, Z. Liu, A. Bair, and J. Z. Kolter (2024) A Simple and Effective Pruning Approach for Large Language Models. In International Conference on Learning Representations (ICLR), 外部链接：链接
  346. [346]R. S. Sutton and A. G. Barto (2018) Reinforcement Learning: An Introduction. 外部链接：链接
  347. [347]R. S. Sutton (1988) Learning to Predict by the Methods of Temporal Differences. Machine Learning.
  348. [348]M. P. Team (2025) TorchForge: PyTorch-Native Post-Training at Scale. 外部链接：链接
  349. [349]P. Team (2024) Torchtune: PyTorch Native Post-Training Library. 外部链接：链接
  350. [350]Q. Team (2024) Qwen2.5: A Party of Foundation Models. arXiv Preprint arXiv:2412.15115. 外部链接：链接
  351. [351]Q. Team (2024) QwQ: Reflect Deeply on the Boundaries of the Unknown. Qwen Blog. 外部链接：链接
  352. [352]Q. Team (2025) Qwen3 Technical Report. arXiv Preprint arXiv:2505.09388.
  353. [353]A. Templeton, T. Conerly, J. Marcus, et al. (2024) Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet. Transformer Circuits Thread. 外部链接：链接
  354. [354]E. Todd, M. L. Li, A. S. Sharma, A. Mueller, B. C. Wallace, and D. Bau (2024) Function Vectors in Large Language Models. In Proceedings of the 12th International Conference on Learning Representations (ICLR), 外部链接：链接
  355. [355]M. Towers, A. Kwiatkowski, J. Terry, et al. (2024) Gymnasium: A Standard Interface for Reinforcement Learning Environments. NeurIPS Datasets and Benchmarks. 外部链接：链接
  356. [356]E. Tulving (1985) Memory and Consciousness. Canadian Psychology / Psychologie Canadienne.
  357. [357]A. Vaswani, N. Shazeer, N. Parmar, et al. (2017) Attention Is All You Need. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  358. [358]Vercel (2024) Vercel AI SDK. 外部链接：链接
  359. [359]J. Vig (2019) A Multiscale Visualization of Attention in the Transformer Model. In Proceedings of the 57th ACL: System Demonstrations, 外部链接：链接
  360. [360]A. K. Vijayakumar, M. Cogswell, R. R. Selvaraju, et al. (2018) Diverse Beam Search: Decoding Diverse Solutions from Neural Sequence Models. AAAI.
  361. [361]E. Voita, D. Talbot, F. Moiseev, R. Sennrich, and I. Titov (2019) Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL), 外部链接：链接
  362. [362]G. Wang, Y. Xie, Y. Jiang, et al. (2023) Voyager: An Open-Ended Embodied Agent with Large Language Models. arXiv Preprint arXiv:2305.16291. 外部链接：链接
  363. [363]J. Wang, J. Duan, Y. Liu, Z. Yue, H. Tong, and J. Wang (2024) Beyond Reverse KL: Generalizing Direct Preference Optimization with Diverse Divergence Constraints. In Proceedings of the 12th International Conference on Learning Representations (ICLR), 外部链接：链接
  364. [364]L. Wang, W. Xu, Y. Lan, et al. (2023) Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL), 外部链接：链接
  365. [365]P. Wang, L. Li, Z. Shao, et al. (2024) Math-Shepherd: Verify and Reinforce LLMs Step-by-Step Without Human Annotations. arXiv Preprint arXiv:2312.08935. 外部链接：链接
  366. [366]X. Wang, B. Ding, Z. Hoang, et al. (2024) OpenHands: An Open Platform for AI Software Developers as Generalist Agents. arXiv Preprint arXiv:2407.16741.
  367. [367]X. Wang, J. Wei, D. Schuurmans, Q. Le, et al. (2023) Self-Consistency Improves Chain of Thought Reasoning in Language Models. In Proceedings of the 11th International Conference on Learning Representations (ICLR), 外部链接：链接
  368. [368]X. Wang, J. Wei, D. Schuurmans, Q. V. Le, et al. (2023) Self-Consistency Improves Chain of Thought Reasoning in Language Models. In International Conference on Learning Representations (ICLR), 外部链接：链接
  369. [369]Y. Wang et al. (2025) RLFactory: A Plug-and-Play Framework for Multi-Round Tool-Use RL. Note: arXiv preprint
  370. [370]Y. e. a. Wang (2025) Improving Data Efficiency for LLM Reinforcement Fine-Tuning via Difficulty-Targeted Online Data Selection. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  371. [371]Y. Wang, Y. Kordi, S. Mishra, et al. (2022) Self-Instruct: Aligning Language Models with Self-Generated Instructions. arXiv Preprint arXiv:2212.10560. 外部链接：链接
  372. [372]Z. Wang et al. (2024) Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts. Note: arXiv preprint
  373. [373]C. J. C. H. Watkins (1989) Learning from Delayed Rewards.
  374. [374]J. Wei, X. Wang, D. Schuurmans, et al. (2022) Chain-of-Thought Prompting Elicits Reasoning in Large Language Models. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  375. [375]G. Weiss and ed (1999) Multiagent Systems: A Modern Approach to Distributed Artificial Intelligence.
  376. [376]L. v. Werra, Y. Belkada, L. Tunstall, et al. (2022) TRL: Transformer Reinforcement Learning. 外部链接：链接
  377. [377]J. Weston, S. Chopra, and A. Bordes (2015) Memory Networks. In International Conference on Learning Representations (ICLR), 外部链接：链接
  378. [378]B. T. Willard and R. Louf (2023) Efficient Guided Generation for Large Language Models. arXiv Preprint arXiv:2307.09702. 外部链接：链接
  379. [379]R. J. Williams (1992) Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning. Machine Learning. 外部链接：链接
  380. [380]E. B. Wilson (1927) Probable Inference, the Law of Succession, and Statistical Inference. Journal of the American Statistical Association. 外部链接：链接
  381. [381]M. Wooldridge, N. R. Jennings, and D. Kinny (2000) The Gaia Methodology for Agent-Oriented Analysis and Design. Autonomous Agents and Multi-Agent Systems.
  382. [382]M. Wooldridge (2009) An Introduction to MultiAgent Systems.
  383. [383]B. Workshop (2023) BLOOM: A 176B-Parameter Open-Access Multilingual Language Model. arXiv Preprint arXiv:2211.05100. 外部链接：链接
  384. [384]D. Wu, H. Wang, W. Yu, Y. Zhang, K. Chang, and D. Yu (2025) LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory. In International Conference on Learning Representations (ICLR), 外部链接：链接
  385. [385]Q. Wu, G. Bansal, J. Zhang, et al. (2023) AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation. arXiv Preprint arXiv:2308.08155.
  386. [386]Y. Wu et al. (2026) Long-Horizon-Terminal-Bench: Dense Reward Grading for Command-Line Agents. Note: arXiv preprint
  387. [387]Z. Wu, A. Arora, Z. Wang, B. Kim, and T. Huang (2024) Retrieval Head Mechanistically Explains Long-Context Factuality. arXiv Preprint arXiv:2404.15574. 外部链接：链接
  388. [388]Z. Wu, Q. Hu, Y. Zhang, Y. Gao, and J. Chen (2024) An Empirical Analysis of Compute-Optimal Inference for Problem-Solving with Language Models. arXiv Preprint arXiv:2408.00724.
  389. [389]xAI (2026) Grok 4.5: Training on Real Developer Session Traces. Note: Technical report
  390. [390]Z. Xi, Y. Ding, W. Chen, et al. (2024) AgentGym: Evolving Large Language Model-Based Agents Across Diverse Environments. arXiv Preprint arXiv:2406.04151. 外部链接：链接
  391. [391]F. Xia, T. Liu, J. Wang, W. Zhang, and H. Li (2008) Listwise Approach to Learning to Rank: Theory and Algorithm. In Proceedings of the 25th International Conference on Machine Learning (ICML),
  392. [392]G. Xiao, J. Lin, M. Seznec, H. Wu, J. Demouth, and S. Han (2023) SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models. In Proceedings of the 40th International Conference on Machine Learning (ICML), 外部链接：链接
  393. [393]G. Xiao, Y. Tian, B. Chen, S. Han, and M. Lewis (2024) Efficient Streaming Language Models with Attention Sinks. In Proceedings of the 12th International Conference on Learning Representations (ICLR), 外部链接：链接
  394. [394]G. Xiao, Y. Tian, B. Chen, S. Han, and M. Lewis (2024) Efficient Streaming Language Models with Attention Sinks. In International Conference on Learning Representations (ICLR), 外部链接：链接
  395. [395]S. Xiao, Z. Liu, P. Zhang, and N. Muennighoff (2023) C-Pack: Packaged Resources to Advance General Chinese Embedding. arXiv Preprint arXiv:2309.07597.
  396. [396]S. M. Xie, A. Raghunathan, P. Liang, and T. Ma (2022) An Explanation of in-Context Learning as Implicit Bayesian Inference. In Proceedings of the 10th International Conference on Learning Representations (ICLR), 外部链接：链接
  397. [397]T. Xie, D. Zhang, J. Chen, et al. (2024) OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments. In Advances in Neural Information Processing Systems,
  398. [398]C. Xu, Q. Sun, K. Zheng, et al. (2023) WizardLM: Empowering Large Language Models to Follow Complex Instructions. arXiv Preprint arXiv:2304.12244. 外部链接：链接
  399. [399]H. Xu, H. Zhao, Y. Liu, and D. Wei (2025) It Takes Two: Pairwise Preference Optimization with Two Rollouts. arXiv Preprint arXiv:2505.07856. 外部链接：链接
  400. [400]W. Xu, Z. Liang, K. Mei, H. Gao, J. Tan, and Y. Zhang (2025) A-Mem: Agentic Memory for LLM Agents. In Advances in Neural Information Processing Systems (NeurIPS),
  401. [401]S. Yan, J. Gu, Y. Zhu, and Z. Ling (2024) Corrective Retrieval Augmented Generation. arXiv Preprint arXiv:2401.15884. 外部链接：链接
  402. [402]C. Yang, X. Wang, Y. Lu, et al. (2024) Large Language Models as Optimizers. In Proceedings of the 12th International Conference on Learning Representations (ICLR), 外部链接：链接
  403. [403]J. Yang, Y. Zhu, B. Wang, et al. (2025) ARQ: Attentive Reasoning Queries for Multi-Hop Question Answering over Long Contexts. arXiv Preprint arXiv:2501.08290. 外部链接：链接
  404. [404]J. Yang, C. E. Jimenez, A. Wettig, et al. (2024) SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  405. [405]Z. Yang, Z. Dai, Y. Yang, J. Carbonell, R. Salakhutdinov, and Q. V. Le (2019) XLNet: Generalized Autoregressive Pretraining for Language Understanding. In Advances in Neural Information Processing Systems (NeurIPS),
  406. [406]Z. Yang, P. Qi, S. Zhang, et al. (2018) HotpotQA: A Dataset for Diverse, Explainable Multi-Hop Question Answering. In Proceedings of EMNLP,
  407. [407]S. Yao, D. Yu, J. Zhao, et al. (2023) Tree of Thoughts: Deliberate Problem Solving with Large Language Models. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  408. [408]S. Yao, D. Yu, J. Zhao, et al. (2024) Tree of Thoughts: Deliberate Problem Solving with Large Language Models. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  409. [409]S. Yao, J. Zhao, D. Yu, et al. (2023) ReAct: Synergizing Reasoning and Acting in Language Models. In International Conference on Learning Representations (ICLR), 外部链接：链接
  410. [410]Z. Yao, S. Rajbhandari, R. Y. Aminabadi, et al. (2023) DeepSpeed-Chat: Easy, Fast and Affordable RLHF Training of ChatGPT-Like Models at All Scales. arXiv Preprint arXiv:2308.01320.
  411. [411]J. Yoo and D. Shin (2025) Adaptive Episode Length Adjustment for Multi-Agent Reinforcement Learning. In Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS),
  412. [412]G. Yu, J. S. Jeong, G. Kim, S. Kim, and B. Chun (2022) Orca: A Distributed Serving System for Transformer-Based Generative Models. In Proceedings of the 16th USENIX Symposium on Operating Systems Design and Implementation (OSDI),
  413. [413]Q. Yu, Z. Sun, S. Wen, et al. (2025) DAPO: An Open-Source LLM Reinforcement Learning System. arXiv Preprint arXiv:2503.14476. 外部链接：链接
  414. [414]X. Yu, B. Peng, R. Xu, et al. (2026) Reinforcement World Model Learning for LLM-Based Agents. arXiv Preprint arXiv:2602.05842.
  415. [415]T. Zadouri, J. Shah, G. Bikshandi, and T. Dao (2026) FlashAttention-4: Hardware-Efficient Attention on Blackwell GPUs with Minimal Software Design. arXiv Preprint arXiv:2603.05451. 外部链接：链接
  416. [416]M. Zaheer, G. Guruganesh, K. A. Dubey, et al. (2020) Big Bird: Transformers for Longer Sequences. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  417. [417]E. B. Zaken, S. Ravfogel, and Y. Goldberg (2022) BitFit: Simple Parameter-Efficient Fine-Tuning for Transformer-Based Masked Language-Models. In Proceedings of the 60th Annual Meeting of the ACL, 外部链接：链接
  418. [418]E. Zelikman, G. Harik, Y. Shao, V. Jayasiri, N. Haber, and N. D. Goodman (2024) Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking. arXiv Preprint arXiv:2403.09629. 外部链接：链接
  419. [419]E. Zelikman, Y. Wu, J. Mu, and N. D. Goodman (2022) STaR: Bootstrapping Reasoning with Reasoning. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  420. [420]A. L. Zhang, S. H. Mahdavi, P. Liang, and T. Hashimoto (2025) Recursive Language Models. arXiv Preprint arXiv:2512.24601. 外部链接：链接
  421. [421]B. Zhang and R. Sennrich (2019) Root Mean Square Layer Normalization. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  422. [422]J. Zhang, S. Yu, D. Chong, et al. (2025) Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversity. arXiv Preprint arXiv:2510.01171. 外部链接：链接
  423. [423]K. Zhang, Z. Yang, and T. Başar (2021) Multi-Agent Reinforcement Learning: A Selective Overview of Theories and Algorithms. Handbook of Reinforcement Learning and Control. 外部链接：链接
  424. [424]Q. Zhang, M. Chen, A. Bukharin, et al. (2023) AdaLoRA: Adaptive Budget Allocation for Parameter-Efficient Fine-Tuning. arXiv Preprint arXiv:2303.10512. 外部链接：链接
  425. [425]T. Zhang, S. G. Patil, N. Jain, et al. (2024) RAFT: Adapting Language Model to Domain Specific RAG. arXiv Preprint arXiv:2403.10131. 外部链接：链接
  426. [426]T. Zhang, V. Kishore, F. Wu, K. Q. Weinberger, and Y. Artzi (2020) BERTScore: Evaluating Text Generation with BERT. In International Conference on Learning Representations (ICLR), 外部链接：链接
  427. [427]W. Zhang et al. (2026) Self-Evolving Software Agents: Combining BDI Architectures with LLMs. Note: arXiv preprint
  428. [428]X. Zhang, Y. Chen, S. Hu, et al. (2024) InfiniteBench: Extending Long Context Evaluation Beyond 100K Tokens. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL), 外部链接：链接
  429. [429]Z. Zhang, Y. Sheng, T. Zhou, et al. (2023) H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  430. [430]S. Zhao, Z. Xie, M. Liu, J. Huang, G. Pang, F. Chen, and A. Grover (2026) Self-Distilled Reasoner: On-Policy Self-Distillation for Large Language Models. In Proceedings of ICLR 2026, Note: arXiv preprint arXiv:2601.18734 外部链接：链接
  431. [431]Y. Zhao, A. Gu, R. Varma, et al. (2023) PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel. In Proceedings of the VLDB Endowment,
  432. [432]Y. Zhao, R. Joshi, T. Liu, M. Khalman, M. Saleh, and P. J. Liu (2023) SLiC-HF: Sequence Likelihood Calibration with Human Feedback. arXiv Preprint arXiv:2305.10425. 外部链接：链接
  433. [433]L. Zheng, W. Chiang, Y. Sheng, et al. (2023) Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  434. [434]Zhipu AI (2025) GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models. arXiv preprint arXiv:2508.06471. 外部链接：链接
  435. [435]Zhipu AI (2026) GLM-5: From Vibe Coding to Agentic Engineering. arXiv preprint arXiv:2602.15763. 外部链接：链接
  436. [436]Y. Zhong, Y. Chen, Z. Li, and M. Chen (2025) Importance Sampling Corrections for Large Language Model Alignment with Asynchronous Generation. arXiv Preprint arXiv:2503.09057. 外部链接：链接
  437. [437]Y. Zhong, J. Shi, Y. Chen, and M. Chen (2025) GDPO: Learning to Directly Align Language Models with Group-Decoupled Reward. arXiv Preprint arXiv:2501.17888. 外部链接：链接
  438. [438]A. Zhou, K. Yan, M. Shlapentokh-Rothman, H. Wang, and Y. Wang (2024) Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models. arXiv Preprint arXiv:2310.04406.
  439. [439]C. Zhou, P. Liu, P. Xu, et al. (2023) LIMA: Less Is More for Alignment. In Advances in Neural Information Processing Systems (NeurIPS), 外部链接：链接
  440. [440]S. Zhou, F. F. Xu, H. Zhu, et al. (2024) WebArena: A Realistic Web Environment for Building Autonomous Agents. In International Conference on Learning Representations (ICLR), 外部链接：链接
  441. [441]Y. Zhou, A. I. Muresanu, Z. Han, et al. (2023) Large Language Models Are Human-Level Prompt Engineers. In Proceedings of the 11th International Conference on Learning Representations (ICLR), 外部链接：链接
  442. [442]D. M. Ziegler, N. Stiennon, J. Wu, et al. (2019) Fine-Tuning Language Models from Human Preferences. arXiv Preprint arXiv:1909.08593. 外部链接：链接
  443. [443]A. Zou, Z. Wang, J. Z. Kolter, and M. Fredrikson (2023) Universal and Transferable Adversarial Attacks on Aligned Language Models. arXiv Preprint arXiv:2307.15043. 外部链接：链接