References (seed bibliography)

Only sources we intend to cite. Items marked [VERIFY] need URL/edition check before chapter publication. Never invent citations.

Models, sampling, context

Vaswani et al., 2017. Attention Is All You Need. arXiv:1706.03762
Liu et al., 2023. Lost in the Middle: How Language Models Use Long Contexts. arXiv:2307.03172
Wei et al., 2022. Chain-of-Thought Prompting Elicits Reasoning in Large Language Models. arXiv:2201.11903
Holtzman et al., 2020. The Curious Case of Neural Text Degeneration. arXiv:1904.09751

Agents & tools

Yao et al., 2022. ReAct: Synergizing Reasoning and Acting in Language Models. arXiv:2210.03629
Anthropic. Building Effective Agents. Engineering publication, 2024. [VERIFY URL]
Shinn et al., 2023. Reflexion. arXiv:2303.11366

Memory

Packer et al. / MemGPT line. MemGPT: Towards LLMs as Operating Systems. [VERIFY exact citation]
Survey literature on cognitive memory types for agents — select carefully; [VERIFY]

Serving, caches, systems

Kwon et al., 2023. Efficient Memory Management for LLM Serving with PagedAttention (vLLM). SOSP.
Zheng et al., 2024. SGLang: Efficient Execution of Structured Language Model Programs. arXiv:2312.07104
Vendor docs: prompt caching / prefix caching (OpenAI, Anthropic, etc.). [VERIFY per vendor]

Distributed systems & logs

Lamport, 1978. Time, Clocks, and the Ordering of Events in a Distributed System. CACM.
Kleppmann, 2017. Designing Data-Intensive Applications. O’Reilly.

Protocols

Model Context Protocol — official documentation. https://modelcontextprotocol.io [VERIFY]
Agent2Agent (A2A) — official specification / docs. [VERIFY URL]

Security

OWASP Top 10 for LLM Applications. [VERIFY edition]
Indirect prompt injection literature (e.g. Greshake et al. and follow-ons). [VERIFY]

Evaluation

Zheng et al. Judging LLM-as-a-Judge… (MT-Bench / Chatbot Arena line). [VERIFY]
Agent evaluation surveys / benchmarks — cite specifically per claim; [VERIFY]

Governance & risk

Board of Governors of the Federal Reserve System. SR 11-7: Guidance on Model Risk Management. 2011.
EU AI Act — official text for high-level mapping only; [VERIFY] jurisdiction applicability
ISO/IEC 42001 — AI management systems; [VERIFY] when citing requirements

Durable execution / workflows

Temporal documentation (concepts: workflows, activities, determinism). [VERIFY]
Cloud workflow engine docs as emerging practice — label as such

Math / control (light touch)

Sutton & Barto. Reinforcement Learning: An Introduction — selected MDP/POMDP sections. [VERIFY edition]
Pearl. Causal Inference primers — only for Ch 44 intuition; [VERIFY]

How to add a source

Add here with stable identifier (arXiv, DOI, official URL).
In chapter, cite sparingly and state what claim it supports.
If you cannot verify, use [VERIFY SOURCE] inline and leave a note in the PR.