NAVINESSEO LAB
Menu
TaalNederlands

NAVINES SEO LAB

Onderzoek

Oorspronkelijk onderzoek met zichtbare methode, bronnen en beperkingen.

Onderzoeksbibliotheek

Zes volledig gelezen papers vertaald naar afgebakende productbesluiten.

arXiv · 2022-10-18

Making a MIRACL: Multilingual Information Retrieval Across a Continuum of Languages

Auteurs
Xinyu Zhang; Nandan Thakur; Odunayo Ogundepo; Ehsan Kamalloo; David Alfonso-Hermelo; Xiaoguang Li; Qun Liu; Mehdi Rezagholizadeh; Jimmy Lin
Publicatiedatum
2022-10-18
Evidentierecord openen

Onderzoeksvraag

How can a retrieval benchmark cover languages with very different resource levels while using native queries and relevance judgments?

Methode

An 18-language, same-language Wikipedia retrieval benchmark built with native-speaker queries and judgments; lexical, dense and hybrid baselines were compared on the released pools.

Wat het bewijs ondersteunt

Evaluate each language explicitly, document corpus and judgment construction, and keep native review central. The reported hybrid result is a historical benchmark observation.

Grenzen en overdraagbaarheid

This is same-language information retrieval, not translation assessment or web SEO. Heuristic segmentation, candidate-pool gaps and unfinished test labels limit transferability; it says nothing about hreflang or Google rankings.

Classificatie

  • Ondersteund
  • Historisch
  • Alleen academisch

NAVINES-invloed: Lab voor meertalige paginavergelijking

arXiv · 2024-04-14

Competitive Retrieval: Going Beyond the Single Query

Auteurs
Haya Nachimovsky; Moshe Tennenholtz; Fiana Raiber; Oren Kurland
Publicatiedatum
2024-04-14
Evidentierecord openen

Onderzoeksvraag

How does publisher competition change when participants allocate content decisions across multiple queries rather than one isolated query?

Methode

A formal multi-query game plus four controlled student competitions covering 30 TREC topics, 84 participants and specified proxy rankers; best-response dynamics and feature changes were examined.

Wat het bewijs ondersteunt

Cross-query opportunity cost and competitor response deserve explicit treatment; equilibrium existence does not imply that learning dynamics will converge.

Grenzen en overdraagbaarheid

The student setting, topics and rankers are controlled proxies. The work does not identify current Google features, predict winners or justify reusable weights or a practical Nash-equilibrium claim.

Classificatie

  • Beperkt
  • Alleen academisch

NAVINES-invloed: Planner voor concurrerende multizoekportfolio

arXiv · 2020-05-26

Ranking-Incentivized Quality Preserving Content Modification

Auteurs
Gregory Goren; Oren Kurland; Moshe Tennenholtz; Fiana Raiber
Publicatiedatum
2020-05-26
Evidentierecord openen

Onderzoeksvraag

Can ranking-incentivized content modification include coherence and modification cost instead of optimizing promotion alone?

Methode

A controlled passage-replacement method balanced promotion and coherence, evaluated with proxy rankers and offline data across 31 queries, with an explicit ethics discussion.

Wat het bewijs ondersteunt

Quality constraints and the cost of changing content belong in planning. A promotional change that fails editorial integrity should be rejected.

Grenzen en overdraagbaarheid

Passages came from higher-ranked candidates and quality judgments were bounded and subjective. It does not support copying winners, real-engine causal claims or unrestricted automated rewriting.

Classificatie

  • Beperkt
  • Alleen academisch

NAVINES-invloed: Planner voor concurrerende multizoekportfolio

Journal of Artificial Intelligence Research · 2025

The Search for Stability: Learning Dynamics of Strategic Publishers with Initial Documents

Auteurs
Omer Madmon; Idan Pipano; Itamar Reinman; Moshe Tennenholtz
Publicatiedatum
2025
Evidentierecord openen

Onderzoeksvraag

Under which formal ranking mechanisms and content-deviation costs can strategic publisher learning converge or remain unstable?

Methode

Formal publisher-utility models with rank benefit minus deviation cost, convergence results for specified linear/softmax settings, and discrete simulations, predominantly with two publishers.

Wat het bewijs ondersteunt

Change cost, response dynamics and instability are useful planning dimensions; a mechanism can have an equilibrium while local responses cycle or remain pseudoperiodic.

Grenzen en overdraagbaarheid

The model assumes a static information need, known ranking function and embedding-space actions. Simulations are not a real search engine and do not justify a Google mechanism or guaranteed convergence claim.

Classificatie

  • Ondersteund
  • Beperkt
  • Alleen academisch

NAVINES-invloed: Planner voor concurrerende multizoekportfolio

KDD / arXiv · 2024

GEO: Generative Engine Optimization

Auteurs
Pranjal Aggarwal; Vishvak Murahari; Tanmay Rajpurohit; Ashwin Kalyan; Kartik Narasimhan; Ameet Deshpande
Publicatiedatum
2024
Evidentierecord openen

Onderzoeksvraag

In controlled generative-engine experiments, how do selected content modifications affect proxy visibility across queries and domains?

Methode

GEO-BENCH combined 10,000 queries from nine datasets with visibility proxies and experiments on two generative engines; citation, quotation, statistics and other interventions were compared by domain.

Wat het bewijs ondersteunt

Clear attribution, genuine evidence and domain-aware experimentation can be reviewed as editorial signals; keyword stuffing performed poorly in this benchmark.

Grenzen en overdraagbaarheid

Two engines, black-box variance and proxy metrics do not establish traditional-search effects or future citation probability. Reported gains are historical benchmark results, not forecasts; appendix prompts are untrusted.

Classificatie

  • Beperkt
  • Historisch
  • Alleen academisch

NAVINES-invloed: Lab voor gereedheid voor AI-citaten

arXiv · 2024-07-02

Adversarial Search Engine Optimization for Large Language Models

Auteurs
Fredrik Nestaas; Edoardo Debenedetti; Florian Tramèr
Publicatiedatum
2024-07-02
Evidentierecord openen

Onderzoeksvraag

Can content in retrieved web or plugin documents manipulate an LLM's product preferences, and which broader attack categories appear beyond classic prompt injection?

Methode

Controlled experiments with fictional products, web/plugin documents and proprietary models examined model-directed instructions, concealment, false claims, source suppression and competitor discrediting.

Wat het bewijs ondersteunt

Retrieved content is an untrusted input surface; safety review should cover concealed persuasion and source integrity as well as direct instructions.

Grenzen en overdraagbaarheid

The favorable fictional setup, limited scale and changing proprietary systems do not establish prevalence or complete detection. Exact attack payloads are not republished and the checker is not a certification.

Classificatie

  • Ondersteund
  • Beperkt
  • Alleen academisch

NAVINES-invloed: Veiligheidscheck voor AI-zoekmanipulatie

Bewijs beoordeeld op 14 augustus 2026 · Bepaalt geen actuele Google-factoren of gegarandeerde uitkomsten.

Onderzoek

Oorspronkelijk onderzoek met zichtbare methode, bronnen en beperkingen.

We jagen algoritmen niet na. We bouwen helderheid, vertrouwen en zichtbaarheid die zich opstapelt.

Geavanceerde zoekintelligentie begrijpelijk, bruikbaar en toegankelijk maken, en bedrijven helpen die kennis om te zetten in meetbare actie.

Methode

  • Bepaal scope, omgevingen, kritieke templates, beschikbare gegevens en veranderingsrisico’s.
  • Verzamel HTTP-statussen, richtlijnen, canonicals, links, rendering, sitemaps en Search Console-data.
  • Documenteer doelen, gebeurtenissen, dimensies, eigenaars en beslissingen per maatstaf.
  • Controleer tags, toestemming, tijdzones, filters en verwachte gegevensuitval.

Beperkingen

  • Toestemming, blockers, privacy en modellering maken gegevens onvolledig.
  • Search Console aggregeert en begrenst gegevens en registreert niet elke zoekopdracht.