86 references with evidence ceilings

Type

Local CSV corpus

Date / accessed

2026-08-24 · 2026-08-24

Evidence ceiling

L — proves local corpus state, not the validity of paper claims.

Supports

B1–B7; C1, C4–C8

Verified: 42 records, 87 fields, 42 DOI values, 41 URLs, 41 abstracts, and 42 local PDFs; SHA-256 c149a69140b0dcedc1855f8cbb02fc5cb1e1346470306c76cd68fbafc2e50468. Proposed: canonical reading-list input and claim-level paper queue.

Type

Local normalized JSON

Date / accessed

2026-08-24 · 2026-08-24

Evidence ceiling

L — editorial routing only until each paper is read and verified.

Supports

B1–B7; C8

Verified: machine-readable normalization of the latest CSV. Proposed: generate reference views, tags, module links, and version checks from one structured source.

Type

Local audit

Date / accessed

2026-08-24; route control updated 2026-08-25 · 2026-08-25

Evidence ceiling

L — routing is an editorial control, not human scientific approval or independent reproduction.

Supports

B1–B7; C1–C8

Verified: preserves the initial 42-record triage and five metadata issues. The separate machine-readable sources/paper_route_crosswalk_v1.6.json is authoritative for final 28-chapter monograph, actual 12-chapter Core Notes citations, 16-week course, and evidence-card claim routes: 42 papers, 48 paper–claim edges, 17 unique claims, and three zero-edge cards. Proposed: retain the Markdown table as historical triage; use the crosswalk for v1.6 route checks.

Type

arXiv tutorial / lecture notes

Date / accessed

Submitted 2025-06-02; v3 2026-03-18 · 2026-08-24

Evidence ceiling

B — design/pedagogy reference; the arXiv record is not GEO evidence.

Supports

B8; C8

Verified: tutorial-style notes with a compact title block, parts, appendices, equations, figures, and pedagogical environments. Proposed: emulate information architecture and teaching-role differentiation, not subject matter.

Type

Local TeX source reference

Date / accessed

v3, 2026-03-18 · 2026-08-24

Evidence ceiling

L/B — architecture reference only; reproducible does not mean publication-ready; no text or figure adaptation.

Supports

B8; C8

Verified: 48 safe archive members; modular main.tex, seven parts, five appendices; SHA-256 b69a997d626dcbb94e3b91d83e3077251779f2d9ab767c4ea4209d84ea135fbf. The declared TeX Live 2025/pdflatex path reproduces an 84-page PDF; six page roles were visually sampled. The build remains warning-bearing, untagged, and has blank Title/Author metadata. License is CC BY-NC-ND 4.0. Proposed: transfer only abstract design principles via original prose and original figures, while enforcing stricter GEO release gates.

Type

Official university course

Date / accessed

Spring 2026 · 2026-08-24

Evidence ceiling

B — strong curriculum model; not GEO-specific.

Supports

B2, B8; C1, C3, C8

Verified: a schedule-first course page with lectures, assignments, readings, and recordings. Proposed: use its high-density weekly rhythm, assignment visibility, and prerequisite signaling.

Type

Official university notes

Date / accessed

2026-08-23 edition · 2026-08-24

Evidence ceiling

B — mathematical pedagogy; not evidence about generative search.

Supports

B4, B8; C1, C5, C8

Verified: a 278-page coherent mathematical notes volume. Proposed: benchmark notation discipline, prerequisite refreshers, exercises, and appendix structure.

Type

Subject-published context page

Date / accessed

Current 2026 snapshot · 2026-08-25

Evidence ceiling

B/E — self-description only; it does not verify this course, institutional ownership, staffing, affiliation, partnership, sponsorship, endorsement, or individual scientific claims.

Supports

B1, B6, B7; C1, C7

Verified: the page presents research themes including GEO, agents/online decision-making, trusted AI, and control/optimization. Proposed: treat those claims as dated research-context questions and use restrained visual cues only.

Type

Industry website

Date / accessed

Current 2026 snapshot · 2026-08-24

Evidence ceiling

E — industry framing/design only; no scientific or performance claims.

Supports

B1, B4, B5; C1, C5, C6

Verified: first-party industry framing organized around understanding, growth, and embedding. Proposed: retain the independently rewritten research loop “observe → explain → intervene → validate” and use brand cues sparingly.

Type

Primary paper; KDD 2024

Date / accessed

Submitted 2023-11-16; v3 2024-06-28 · 2026-08-24

Evidence ceiling

A — peer-reviewed, but results are setup-, engine-, metric-, and date-bounded.

Supports

B1, B4, B5; C1, C5, C6

Verified: introduces GEO, GEO-Bench, visibility metrics, and controlled content interventions; reports gains up to 40% in its experimental setting. Proposed: field origin, baseline methods, and a reproduction/critique lab.

Type

Official venue index

Date / accessed

2024 · 2026-08-24

Evidence ceiling

A/B — bibliographic verification, not independent replication.

Supports

B1; C1

Verified: official venue listing for the GEO paper. Proposed: use only to verify publication status and venue metadata.

Type

NeurIPS primary paper

Date / accessed

2020 · 2026-08-24

Evidence ceiling

A — foundational open-system evidence; not a disclosure of current commercial search stacks.

Supports

B2, B3; C1, C3, C4

Verified: formalizes a parametric-plus-non-parametric retrieval/generation architecture and evaluates knowledge-intensive tasks. Proposed: canonical mechanism figure and notation for the retrieval-to-generation pipeline.

Type

EMNLP primary paper

Date / accessed

2020 · 2026-08-24

Evidence ceiling

A — benchmark-bounded; later embedding models require fresh evaluation.

Supports

B2, B4; C3

Verified: evaluates dual-encoder dense retrieval for open-domain QA. Proposed: contrast sparse, dense, and hybrid candidate retrieval in the retrieval lab.

Type

EMNLP primary paper / ALCE benchmark

Date / accessed

2023 · 2026-08-24

Evidence ceiling

A — benchmark and model snapshot bounded.

Supports

B3, B4; C4, C5

Verified: evaluates correctness and completeness of citations for long-form generation. Proposed: distinguish answer quality, citation correctness, citation completeness, and source quality.

Type

NIST evaluation program

Date / accessed

Ongoing; 2026 program visible · 2026-08-24

Evidence ceiling

A/B — authoritative benchmark program; each track’s task and judgments define its ceiling.

Supports

B2–B4; C3–C5, C8

Verified: long-running shared evaluation program; current tracks include retrieval-augmented generation. Proposed: model the course’s evaluation contracts, run files, qrels, and reproducibility package on TREC conventions.

Type

Official benchmark data page

Date / accessed

Created 2025-03-11 for TREC 2024 · 2026-08-24

Evidence ceiling

A — benchmark evidence only; licensing and redistribution must be checked per resource.

Supports

B3, B4; C4, C5, C8

Verified: publishes topics, qrels, nuggets, and citation-support judgments. Proposed: adapt a small lawful subset or analogous original set for answer/citation scoring labs.

Type

Official track overview

Date / accessed

TREC 2025 / published 2026 · 2026-08-24

Evidence ceiling

A — task-specific evaluation design, not a universal quality metric.

Supports

B3, B4; C4, C5

Verified: documents sentence-level answer evaluation and citation-oriented judgments. Proposed: update the rubric beyond citation counts to claim-level support.

Type

Official university course archive

Date / accessed

Spring 2019 · 2026-08-24

Evidence ceiling

B — durable foundations; web-platform details are dated.

Supports

B2, B4, B8; C1–C3

Verified: covers indexing, retrieval, BM25, evaluation/NDCG, crawling, and web search. Proposed: prerequisite bridge and classical-IR comparison boxes.

Type

Academic textbook, free online edition

Date / accessed

Book 2008; online 2009 · 2026-08-24

Evidence ceiling

B — foundational, not current neural/generative-search evidence.

Supports

B2, B4, B8; C1, C3

Verified: canonical treatment of indexing, scoring, evaluation, crawling, and link analysis. Proposed: prerequisite appendix and notation source.

Type

Official university course

Date / accessed

Fall 2026; page updated 2026-05-28 · 2026-08-24

Evidence ceiling

B — current academic syllabus; individual lecture claims need readings.

Supports

B2–B4; C1, C3–C5

Verified: covers text search, RAG, and recommender/search systems in one contemporary course. Proposed: benchmark sequencing from classical retrieval to generative applications.

Type

Official doctoral course

Date / accessed

2025–2026 · 2026-08-24

Evidence ceiling

B — syllabus-level evidence only.

Supports

B2, B6; C1, C3, C8

Verified: combines LLM-based IR, vector search, reranking, conversational search, and query prediction. Proposed: cross-check advanced-course outcomes and workload.

Type

Official university syllabus

Date / accessed

2025–2026 · 2026-08-24

Evidence ceiling

B — syllabus-level evidence only.

Supports

B2, B6; C1, C3

Verified: integrates IR/search with LLM-based conversational agents. Proposed: compare European learning outcomes and assessment balance.

Type

Official project-oriented university course

Date / accessed

Fall 2025 · 2026-08-24

Evidence ceiling

B — pedagogical model; not a guarantee that any method eliminates hallucination.

Supports

B2, B3, B6; C3, C4, C8

Verified: explicitly asks how to perform RAG without hallucination and retrieve from databases/knowledge graphs, with final projects. Proposed: inspiration for the course capstone and evidence-curation studio.

Type

SIGIR 2024 tutorial site

Date / accessed

2024-07-14 · 2026-08-24

Evidence ceiling

B — expert tutorial synthesis; primary papers remain claim authorities.

Supports

B2, B4; C1, C3

Verified: tutorial materials and bibliography on generative retrieval. Proposed: scholarly bridge from document retrieval to generative IR; use slides only within their license.

Type

Open practical course

Date / accessed

2023 · 2026-08-24

Evidence ceiling

B — practical engineering reference; framework and API details age quickly.

Supports

B6, B8; C3, C8

Verified: covers augmented language models, deployment, UX, and LLM operations. Proposed: derive production checklists and end-to-end project rhythm.

Type

Official interactive tutorial / Colab

Date / accessed

Current 2026 snapshot · 2026-08-24

Evidence ceiling

B — runnable tutorial; library behavior and examples can drift.

Supports

B2, B6; C3, C8

Verified: demonstrates query reformulation, decomposition, retrieval, reranking, and multi-step validation. Proposed: optional agentic-retrieval lab with frozen dependencies.

Type

Vendor-hosted educational short course

Date / accessed

2023-11 · 2026-08-24

Evidence ceiling

E/B — instructional value; named metrics and product concepts need independent validation.

Supports

B2–B4; C3, C4

Verified: approximately two hours with videos and code on retrieval evaluation and advanced chunk/retrieval strategies. Proposed: use as optional preparatory practice, not as the course’s evidence authority.

Type

Official research repository; Apache-2.0

Date / accessed

2024 snapshot · 2026-08-24

Evidence ceiling

B/A — strongest implementation companion to A01; code does not independently validate results.

Supports

B4, B5; C5, C6, C8

Verified: paper-linked code, benchmark assets, and evaluation implementation. Proposed: pin a commit, reproduce one bounded baseline, then document deviations and failures.

Type

Official research repository / benchmark

Date / accessed

NeurIPS 2021; release 2.2.0 on 2025-06-04 · 2026-08-24

Evidence ceiling

B/A — reproducible benchmark; dataset licenses and domain fit vary.

Supports

B2, B4; C3, C5, C8

Verified: heterogeneous IR evaluation across multiple datasets and retrieval systems. Proposed: small sparse-vs-dense-vs-hybrid lab and domain-shift discussion.

Type

Maintained academic IR toolkit

Date / accessed

Current 2026 snapshot · 2026-08-24

Evidence ceiling

B — implementation authority for the toolkit, not evidence about closed platforms.

Supports

B2, B4; C3, C8

Verified: reproducible sparse, dense, and hybrid first-stage retrieval over Lucene/Faiss with prebuilt indexes and qrels. Proposed: default retrieval lab toolkit with a pinned environment.

Type

Academic LLM-reranking toolkit

Date / accessed

Current 2026 snapshot · 2026-08-24

Evidence ceiling

B — toolkit and associated-paper evidence only; model/API drift must be frozen.

Supports

B2, B4; C3, C5

Verified: supports reproducible pointwise/listwise LLM reranking workflows. Proposed: compare BM25, dense, and LLM reranking under fixed candidate sets and budgets.

Type

Open benchmark framework

Date / accessed

2022–current · 2026-08-24

Evidence ceiling

B/A — benchmark result ceiling follows dataset coverage and evaluation protocol.

Supports

B2, B4, B6; C3, C5

Verified: standardized evaluation across embedding tasks including retrieval. Proposed: teach model selection as dataset- and language-dependent rather than leaderboard universalism.

Type

Open-source RAG evaluation framework

Date / accessed

Current 2026 snapshot · 2026-08-24

Evidence ceiling

B — framework metrics are operationalizations, not ground truth.

Supports

B3, B4; C4, C5, C8

Verified: implements dataset generation, RAG evaluation, and feedback workflows. Proposed: optional lab scaffold while separately validating metric definitions and judge reliability.

Type

Academic LM-programming and optimization framework

Date / accessed

ICLR 2024 / current · 2026-08-24

Evidence ceiling

B/A — framework is credible; outcomes depend on metric, data, and model.

Supports

B4–B6; C6, C8

Verified: modular LM programs and metric-driven compilation/optimization. Proposed: controlled multi-objective intervention lab with train/dev/test separation.

Type

Open-source evaluation and red-team toolkit

Date / accessed

Current 2026 snapshot · 2026-08-24

Evidence ceiling

B — tooling reference; generated scores require audited assertions and judges.

Supports

B4, B7, B8; C5, C7, C8

Verified: supports test matrices, assertions, red-team checks, and CI workflows. Proposed: capstone regression suite for prompts, answers, citations, and policy checks.

Type

Open-source tracing/evaluation toolkit

Date / accessed

Current 2026 snapshot · 2026-08-24

Evidence ceiling

B — implementation only; metric validity must be independently assessed.

Supports

B3, B4, B8; C4, C5, C8

Verified: supports tracing and feedback functions for retrieval and generation. Proposed: optional observability comparison with Ragas; do not import vendor benchmark claims.

Type

Community proposal / repository

Date / accessed

Proposed 2024-09-03; updated 2026 · 2026-08-24

Evidence ceiling

E — no general ranking/citation claim and no cross-platform requirement.

Supports

B8; C2, C6

Verified: a voluntary proposal for an /llms.txt file, not an IETF/W3C/ISO web standard. Google’s 2026 guide explicitly says Google Search ignores it. Proposed: use only in an experiment contrasting proposals with verified platform behavior.

Type

Official Google Search documentation

Date / accessed

Last updated 2026-07-10 · 2026-08-24

Evidence ceiling

C — authoritative for Google Search only; no cross-engine or outcome guarantee.

Supports

B1, B2, B5, B8; C1, C2, C6

Verified: Google says existing SEO fundamentals apply; describes RAG and query fan-out; states indexing/serving are not guaranteed; says Google Search ignores llms.txt and no special AI markup is required. Proposed: primary platform-specific boundary text and myth-busting lab.

Type

Official Google Search documentation

Date / accessed

Current 2026 snapshot · 2026-08-24

Evidence ceiling

C — Google-specific policy guidance, not a scientific quality metric.

Supports

B5, B7, B8; C6, C7

Verified: mass generation without user value may violate scaled-content-abuse policy; emphasizes accuracy, quality, relevance, and appropriate creation context. Proposed: responsible publishing checklist.

Type

Official Google Search documentation

Date / accessed

Last updated 2026-07-10 · 2026-08-24

Evidence ceiling

C — Google eligibility/policy only.

Supports

B5, B8; C2, C6

Verified: valid structured data creates eligibility for supported features but does not guarantee display. Proposed: schema validation lab with explicit “eligibility ≠ citation causality” language.

Type

Official Microsoft/Bing product guidance

Date / accessed

2026-02-10 · 2026-08-24

Evidence ceiling

C — early-preview, aggregated, Microsoft-surface-specific data.

Supports

B4, B8; C2, C5

Verified: preview reports citations, cited pages, sampled grounding queries, and trends across specified Microsoft surfaces; counts do not indicate rank, authority, or placement. Proposed: measurement-interface case study and metric-definition critique.

Type

Official Microsoft/Bing guidance

Date / accessed

2025-07-31 · 2026-08-24

Evidence ceiling

C — Bing-specific operational guidance, not a citation guarantee.

Supports

B8; C2

Verified: recommends accurate XML sitemaps and lastmod; explicitly says no tool guarantees appearance in AI-generated results. Proposed: sitemap freshness exercise.

Type

Official Microsoft/Bing guidance

Date / accessed

2025-10-15 · 2026-08-24

Evidence ceiling

C — Bing-supported environments only.

Supports

B3, B8; C2, C4

Verified: marked content may remain indexed while being excluded from Bing snippets and AI summaries. Proposed: publisher-control lab contrasting crawl, index, retrieval, and display controls.

Type

Official OpenAI help documentation

Date / accessed

Page reported updated 2026-07-31 · 2026-08-24

Evidence ceiling

C — OpenAI-product-specific and changeable; recheck before publication.

Supports

B2, B8; C2, C5

Verified: distinguishes OAI-SearchBot access for search inclusion from GPTBot training controls and documents ChatGPT referral UTM parameters. Proposed: crawler-policy matrix and log-analysis lab.

Type

Official OpenAI help documentation

Date / accessed

Page reported updated 2026-08-22 · 2026-08-24

Evidence ceiling

C — product behavior can change; not an algorithm specification.

Supports

B1, B3, B4; C1, C4, C5

Verified: describes current web-search behavior, source review, query rewriting, and site eligibility; explicitly says placement is not guaranteed. Proposed: UI/surface observation protocol, with screenshots dated and labeled.

Type

Official Perplexity documentation

Date / accessed

Current 2026 snapshot · 2026-08-24

Evidence ceiling

C — Perplexity-specific; user agents and IP ranges must be revalidated at lab time.

Supports

B2, B8; C2

Verified: distinguishes PerplexityBot and user-triggered retrieval behavior and publishes crawler information. Proposed: first-party source for crawler identification and access-log parsing.

Type

Official Perplexity help documentation

Date / accessed

Updated 2026-07-16 · 2026-08-24

Evidence ceiling

C — stated platform policy, not independent compliance evidence.

Supports

B7, B8; C2, C7

Verified: explains Perplexity’s stated treatment of robots controls. Proposed: compare declared policy with server-log observations without inferring intent from a user agent alone.

Type

IETF Standards Track RFC

Date / accessed

2022-09 · 2026-08-24

Evidence ceiling

N — protocol semantics only; REP is not access authorization and does not guarantee crawler compliance or visibility.

Supports

B8; C2, C7

Verified: normative syntax and matching rules for robots.txt. Proposed: authoritative protocol basis for all crawler-control examples.

Type

Open web protocol documentation

Date / accessed

Current protocol page; n.d. · 2026-08-24

Evidence ceiling

N/B — protocol syntax only; submission does not guarantee crawling, indexing, ranking, or citation.

Supports

B8; C2

Verified: defines sitemap structure and limits. Proposed: technical discoverability lab and validation checklist.

Type

Community vocabulary documentation

Date / accessed

Version 30.0, 2026-03-19 · 2026-08-24

Evidence ceiling

N/B — vocabulary semantics only; no platform visibility effect is implied.

Supports

B5, B8; C2, C6

Verified: documents a shared structured-data vocabulary. Proposed: semantic entity/page representation exercise.

Type

Open URL-notification protocol

Date / accessed

Current 2026 snapshot · 2026-08-24

Evidence ceiling

N/B — notification, not a promise of crawl, index, rank, or citation.

Supports

B8; C2, C5

Verified: specifies how sites notify participating engines of added, updated, or deleted URLs. Proposed: instrument notification latency and downstream crawl observations.

Type

W3C Recommendation

Date / accessed

2023-10-05; updated 2024-12-12 · 2026-08-24

Evidence ceiling

N — accessibility conformance only; no GEO performance claim.

Supports

B8; C2, C8

Verified: normative accessibility success criteria. Proposed: mandatory quality gate for the course site, figures, forms, and labs.

Type

U.S. government voluntary framework

Date / accessed

2023-01-26; revision in progress in 2026 · 2026-08-24

Evidence ceiling

N — voluntary risk framework, not law or certification and not GEO-specific.

Supports

B7, B8; C7, C8

Verified: voluntary framework for incorporating trustworthiness into AI design, development, use, and evaluation. Proposed: structure governance around Govern, Map, Measure, and Manage while flagging the active revision.

Type

U.S. government technical report

Date / accessed

2024-07-26; page updated 2026-04-08 · 2026-08-24

Evidence ceiling

N — voluntary cross-sector guidance; implementations require contextual tailoring.

Supports

B7, B8; C7, C8

Verified: cross-sector companion profile for GenAI risks and actions. Proposed: risk taxonomy for evidence integrity, information security, content provenance, and incident response.

Type

International standard

Date / accessed

2023-12 · 2026-08-24

Evidence ceiling

N — official metadata/overview is public; full requirements are paywalled and must not be inferred from the overview.

Supports

B7, B8; C7, C8

Verified: specifies requirements for establishing, implementing, maintaining, and continually improving an AI management system. Proposed: organizational governance checklist and capstone audit trail.

Type

Industry technical specification

Date / accessed

2.4, 2026-04 · 2026-08-24

Evidence ceiling

N/B — provenance integrity and validation only; not truth verification, authorship proof, or ranking evidence.

Supports

B3, B6–B8; C4, C7, C8

Verified: specifies cryptographically bound, tamper-evident provenance manifests; expressly does not judge whether provenance is “good” or “bad.” Proposed: multimodal provenance and disclosure lab.

Type

PRC regulation / official government text

Date / accessed

Issued 2023-07-10; effective 2023-08-15 · 2026-08-24

Evidence ceiling

N — official legal text; not legal advice and scope/application require counsel.

Supports

B7, B8; C7

Verified: applies to specified public-facing generative AI services in mainland China and sets provider obligations. Proposed: jurisdiction-aware governance unit; obtain legal review for operational advice.

Type

PRC regulatory measure / official text

Date / accessed

Issued 2025-03-07; effective 2025-09-01 · 2026-08-24

Evidence ceiling

N — official legal measure; not legal advice.

Supports

B3, B6–B8; C7, C8

Verified: requires specified explicit and implicit labels and related platform/provider duties. Proposed: publishing and provenance compliance checklist.

Type

Mandatory Chinese national standard

Date / accessed

Published 2025-02-28; effective 2025-09-01 · 2026-08-24

Evidence ceiling

N — normative within scope; no relevance to ranking/citation performance.

Supports

B3, B6–B8; C7, C8

Verified: current mandatory standard for AI-generated/synthetic content labeling methods. Proposed: standards crosswalk with S06 and C2PA, carefully separating mandatory labeling from optional provenance technologies.

Type

Recommended Chinese national standard

Date / accessed

Published 2025-04-25; effective 2025-11-01 · 2026-08-24

Evidence ceiling

N — recommended standard, not law and not a GEO-performance standard.

Supports

B7, B8; C7, C8

Verified: current recommended security standard for generative AI services. Proposed: security-governance matrix for a GEO measurement or optimization service.

Type

Recommended Chinese national standard

Date / accessed

Published 2025-04-25; effective 2025-11-01 · 2026-08-24

Evidence ceiling

N — data-security scope only; not directly a content-ranking rule.

Supports

B7, B8; C7, C8

Verified: current recommended standard covering pre-training and fine-tuning data security. Proposed: data lineage, provenance, licensing, and contamination checklist.

Type

Recommended Chinese national standard

Date / accessed

2020-03-06; effective 2020-10-01; revision in progress in 2026 · 2026-08-24

Evidence ceiling

N — recommended standard and subject to revision; obtain legal/privacy review.

Supports

B4, B7, B8; C5, C7, C8

Verified: current personal-information security specification, with a revision process underway. Proposed: privacy controls for query logs, user studies, analytics, and model outputs.

Type

Chinese standardization guidance project

Date / accessed

Registered 2025-06-20; awaiting approval on 2026-08-24 · 2026-08-24

Evidence ceiling

N-watch — cannot be cited as an in-force standard or requirement.

Supports

B2, B8; C3, C8

Verified: an approval-stage project, not a published national standard. Proposed: standards watchlist only; refresh status before publication.

Type

Chinese standardization project

Date / accessed

2025 project · 2026-08-24

Evidence ceiling

N-watch — project metadata only.

Supports

B2, B6, B8; C3, C8

Verified: registered project concerning Graph RAG, not a published standard. Proposed: horizon note in the agentic/graph retrieval chapter.

Type

Practitioner resource map

Date / accessed

2026-04-24 · 2026-08-24

Evidence ceiling

E — curated secondary source with some broad or time-sensitive claims.

Supports

B1, B8; C1, C8

Verified: role-based map of papers, courses, tools, and implementation topics. Proposed: discovery index and competitor-curriculum checklist; verify every downstream claim at its original source.

Type

Practitioner explainer

Date / accessed

2026-01-18 · 2026-08-24

Evidence ceiling

E — conceptual framing only; funnel claims need independent data.

Supports

B1, B4; C1, C5

Verified: offers a conceptual contrast between link-oriented and synthesized-answer journeys. Proposed: introductory discussion prompt, paired with Google’s first-party guide and measurement evidence.

Type

Secondary paper explainer

Date / accessed

2026-02-17 · 2026-08-24

Evidence ceiling

E — secondary explanation, never the primary citation for paper results.

Supports

B1, B5; C1, C6

Verified: practitioner-oriented explanation of A01. Proposed: optional pre-reading for nontechnical learners; all numerical claims cite A01 instead.

Type

Practitioner implementation tutorial

Date / accessed

2026-02-09 · 2026-08-24

Evidence ceiling

E — bot identities, purposes, and policies are time-sensitive; no compliance assumption.

Supports

B8; C2, C7

Verified: provides configuration examples and crawler taxonomy. Proposed: debugging cases only after rechecking each user agent against vendor docs and RFC 9309.

Type

Practitioner diagnostic taxonomy

Date / accessed

2026-06-23 · 2026-08-24

Evidence ceiling

E — useful hypotheses, not evidence that hidden platform stages work as described.

Supports

B2–B4; C4, C5

Verified: separates retrieval, generation, citation scoring, post-processing, and UI failure hypotheses. Proposed: convert into a falsifiable diagnostic worksheet anchored to ALCE, TREC RAG, and controlled logs.

Type

Practitioner technical tutorial

Date / accessed

2026 · 2026-08-24

Evidence ceiling

E — method inspiration; identity verification must use first-party data and reverse/forward DNS where applicable.

Supports

B4, B8; C2, C5

Verified: presents server-log parsing as an observation method. Proposed: privacy-safe lab distinguishing claimed bot identity, verified IP, request, response, and downstream outcome.

Type

Practitioner research synthesis

Date / accessed

2026-08-18 · 2026-08-24

Evidence ceiling

E — derived synthesis; the decomposition must be formalized and empirically validated.

Supports

B4; C5, C6

Verified: proposes separating within-time repeated-sampling variance from between-time drift. Proposed: motivate repeated measures and time-blocked designs, with claims grounded in the local paper corpus.

Type

Vendor observational study

Date / accessed

2025-05-19; data updated in-page · 2026-08-24

Evidence ceiling

D — descriptive only; proprietary index/parser, logged-out sampling, and no causal inference.

Supports

B4; C5

Verified: reports a very large Ahrefs-owned SERP sample and discloses important coverage limitations. Proposed: example of scale, sampling frames, and why headline percentages need denominators and dates.

Type

Vendor observational study

Date / accessed

2026-03-02 · 2026-08-24

Evidence ceiling

D — Google/Ahrefs sample only; associations do not reveal ranking mechanisms.

Supports

B2, B4; C5

Verified: analyzes 863,000 SERPs and four million AI Overview URLs with an updated parser. Proposed: platform-drift case study and replication-design prompt.

Type

Vendor matched difference-in-differences study

Date / accessed

2026-05-11 · 2026-08-24

Evidence ceiling

D — stronger quasi-experimental design, but proprietary data, treatment timing uncertainty, and residual confounding remain.

Supports

B4, B5; C5, C6

Verified: tracks 1,885 treated and 4,000 control pages and reports no major citation uplift across tested surfaces. Proposed: teach correlation vs treatment effects and challenge “schema guarantees citations.”

Type

Vendor observational study

Date / accessed

2026-06-09 · 2026-08-24

Evidence ceiling

D — small prompt set and proprietary sampling; descriptive only.

Supports

B3, B4; C4, C5

Verified: distinguishes cited and mentioned outcomes using 3,981 domain appearances, 115 prompts, 14 countries, and four named surfaces. Proposed: metric-separation exercise, not a population estimate.

Type

Vendor longitudinal observational study

Date / accessed

2025 · 2026-08-24

Evidence ceiling

D — proprietary sampling/parser and Google-specific scope.

Supports

B4; C5

Verified: reports patterns over more than 200,000 keywords during 2025. Proposed: discuss longitudinal sampling and platform-change confounding.

Type

Vendor report hub

Date / accessed

Current 2026 snapshot · 2026-08-24

Evidence ceiling

D/E — hub metadata only; report-specific evidence ceilings unknown until review.

Supports

B1, B4; C5

Verified: lists current AI-citation and visibility reports. Proposed: horizon-scanning queue; do not quote a report until its full methodology is audited.

Type

Vendor analytics report / PDF

Date / accessed

2025 · 2026-08-24

Evidence ceiling

D — customer/sample and attribution-method limits; traffic is not citation visibility.

Supports

B1, B4; C5

Verified: Adobe analytics report about AI-referred traffic and behavior. Proposed: contrast referral traffic with answer visibility, citation, and mention metrics.

Type

Vendor research report / PDF

Date / accessed

2025-05 · 2026-08-24

Evidence ceiling

D — proprietary parser/sample, commercial framing, and no independent replication.

Supports

B4, B5; C5

Verified: ten-page BrightEdge report based on its Generative Parser and dated one year after the AIO launch. Proposed: compare metric definitions and apparently conflicting vendor estimates; respect the report’s “not for distribution without consent” notice.

Type

Vendor industry report / PDF

Date / accessed

2025-09 · 2026-08-24

Evidence ceiling

D — proprietary sample and attribution; referral traffic does not measure answer visibility or causal influence.

Supports

B1, B4; C5

Verified: eight-page report analyzing BrightEdge customer/query data from January–August 2025 and separating AI referrals from organic search. Proposed: traffic-attribution case study and a prompt to audit denominator, channel definition, and conversion window.

Type

Official university lecture playlist

Date / accessed

2025 · 2026-08-24

Evidence ceiling

B — instructional recording; cite notes/papers for claims.

Supports

B2; C1

Verified: recorded course sequence from foundations through training/evaluation topics. Proposed: prerequisite clips and lecture-pacing reference.

Type

Official university seminar recordings

Date / accessed

Multi-year; current 2026 page · 2026-08-24

Evidence ceiling

B — expert talks, not peer review.

Supports

B2, B6; C1, C3

Verified: includes “Retrieval Augmented Language Models” by Douwe Kiela among expert talks. Proposed: optional conceptual viewing with guided questions.

Type

Official conference keynote video

Date / accessed

2024-07-15 · 2026-08-24

Evidence ceiling

B — expert synthesis; underlying papers support technical claims.

Supports

B2, B4; C1, C3

Verified: Yiming Yang surveys representation learning, dense IR, knowledge-enhanced retrieval, and RAG-related work. Proposed: field-history and research-frontier discussion.

Type

Practical course lecture video

Date / accessed

2023-05-11 · 2026-08-24

Evidence ceiling

B/E — strong engineering pedagogy; examples and APIs are dated.

Supports

B2, B6, B8; C1, C3

Verified: explains retrieval, chaining, tools, and augmented-LM application structure. Proposed: systems overview before the implementation lab.

Type

Practitioner community event recording

Date / accessed

2026-07-31 · 2026-08-24

Evidence ceiling

E — event discussion only.

Supports

B1; C1

Verified: live practitioner/community event connected to The GEO Community. Proposed: capture contemporary terminology and practitioner questions, not technical evidence.

Type

Official university lecture video

Date / accessed

2025 · 2026-08-24

Evidence ceiling

B — pedagogy and prerequisite framing only.

Supports

B8; C1

Verified: a clear example of opening a technically dense course with objectives, system scope, and implementation expectations. Proposed: production reference for the GEO course trailer/first lecture.