NAVINESSEO LAB
Menu
LinguaItaliano

NAVINES SEO LAB

Ricerca

Studi originali con metodo, fonti e limiti espliciti.

Biblioteca delle evidenze

Sei articoli letti integralmente e tradotti in decisioni di prodotto delimitate.

arXiv · 2022-10-18

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

Autori
Xinyu Zhang; Nandan Thakur; Odunayo Ogundepo; Ehsan Kamalloo; David Alfonso-Hermelo; Xiaoguang Li; Qun Liu; Mehdi Rezagholizadeh; Jimmy Lin
Data di pubblicazione
2022-10-18
Apri scheda delle prove

Domanda di ricerca

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

Metodo

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.

Cosa sostengono le prove

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

Limiti e trasferibilità

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.

Classificazione

  • Supportato
  • Storico
  • Solo accademico

Influenza NAVINES: Laboratorio di confronto di pagine multilingue

arXiv · 2024-04-14

Competitive Retrieval: Going Beyond the Single Query

Autori
Haya Nachimovsky; Moshe Tennenholtz; Fiana Raiber; Oren Kurland
Data di pubblicazione
2024-04-14
Apri scheda delle prove

Domanda di ricerca

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

Metodo

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.

Cosa sostengono le prove

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

Limiti e trasferibilità

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.

Classificazione

  • Limitato
  • Solo accademico

Influenza NAVINES: Pianificatore di portafoglio competitivo multiquery

arXiv · 2020-05-26

Ranking-Incentivized Quality Preserving Content Modification

Autori
Gregory Goren; Oren Kurland; Moshe Tennenholtz; Fiana Raiber
Data di pubblicazione
2020-05-26
Apri scheda delle prove

Domanda di ricerca

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

Metodo

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

Cosa sostengono le prove

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

Limiti e trasferibilità

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.

Classificazione

  • Limitato
  • Solo accademico

Influenza NAVINES: Pianificatore di portafoglio competitivo multiquery

Journal of Artificial Intelligence Research · 2025

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

Autori
Omer Madmon; Idan Pipano; Itamar Reinman; Moshe Tennenholtz
Data di pubblicazione
2025
Apri scheda delle prove

Domanda di ricerca

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

Metodo

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

Cosa sostengono le prove

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

Limiti e trasferibilità

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.

Classificazione

  • Supportato
  • Limitato
  • Solo accademico

Influenza NAVINES: Pianificatore di portafoglio competitivo multiquery

KDD / arXiv · 2024

GEO: Generative Engine Optimization

Autori
Pranjal Aggarwal; Vishvak Murahari; Tanmay Rajpurohit; Ashwin Kalyan; Kartik Narasimhan; Ameet Deshpande
Data di pubblicazione
2024
Apri scheda delle prove

Domanda di ricerca

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

Metodo

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.

Cosa sostengono le prove

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

Limiti e trasferibilità

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.

Classificazione

  • Limitato
  • Storico
  • Solo accademico

Influenza NAVINES: Laboratorio di preparazione alle citazioni IA

arXiv · 2024-07-02

Adversarial Search Engine Optimization for Large Language Models

Autori
Fredrik Nestaas; Edoardo Debenedetti; Florian Tramèr
Data di pubblicazione
2024-07-02
Apri scheda delle prove

Domanda di ricerca

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

Metodo

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

Cosa sostengono le prove

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

Limiti e trasferibilità

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.

Classificazione

  • Supportato
  • Limitato
  • Solo accademico

Influenza NAVINES: Controllo di sicurezza contro la manipolazione IA

Evidenza rivista il 14 agosto 2026 · Non stabilisce fattori Google attuali né risultati garantiti.

Ricerca

Studi originali con metodo, fonti e limiti espliciti.

Non inseguiamo gli algoritmi. Costruiamo chiarezza, fiducia e visibilità che cresce nel tempo.

Rendere l’intelligenza avanzata della ricerca comprensibile, utile e accessibile, aiutando poi le aziende a trasformarla in azioni misurabili.

Metodo

  • Definisci ambito, ambienti, template critici, dati disponibili e rischi del cambiamento.
  • Raccogli stati HTTP, direttive, canonical, link, rendering, sitemap e dati Search Console.
  • Documenta obiettivi, eventi, dimensioni, responsabili e decisioni legate a ogni metrica.
  • Verifica tag, consenso, fusi orari, filtri e perdite di dati previste.

Limiti

  • Consenso, blocchi, privacy e modellazione rendono i dati incompleti.
  • Search Console aggrega e limita i dati; non registra ogni singola ricerca.