Knižnica výskumných dôkazov

Šesť prác bolo prečítaných v plnom znení a prevedených na ohraničené produktové rozhodnutia.

arXiv · 2022-10-18

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

Otvoriť záznam dôkazov

Výskumná otázka

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

Metóda

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.

Čo dôkazy podporujú

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

Limity a prenosnosť

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.

Klasifikácia

  • Podporené
  • Historické
  • Len akademické

Vplyv na NAVINES: Laboratórium porovnania viacjazyčných stránok

arXiv · 2024-04-14

Competitive Retrieval: Going Beyond the Single Query

Otvoriť záznam dôkazov

Výskumná otázka

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

Metóda

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.

Čo dôkazy podporujú

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

Limity a prenosnosť

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.

Klasifikácia

  • Obmedzené
  • Len akademické

Vplyv na NAVINES: Plánovač konkurenčného portfólia dotazov

arXiv · 2020-05-26

Ranking-Incentivized Quality Preserving Content Modification

Otvoriť záznam dôkazov

Výskumná otázka

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

Metóda

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

Čo dôkazy podporujú

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

Limity a prenosnosť

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.

Klasifikácia

  • Obmedzené
  • Len akademické

Vplyv na NAVINES: Plánovač konkurenčného portfólia dotazov

Journal of Artificial Intelligence Research · 2025

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

Otvoriť záznam dôkazov

Výskumná otázka

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

Metóda

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

Čo dôkazy podporujú

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

Limity a prenosnosť

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.

Klasifikácia

  • Podporené
  • Obmedzené
  • Len akademické

Vplyv na NAVINES: Plánovač konkurenčného portfólia dotazov

KDD / arXiv · 2024

GEO: Generative Engine Optimization

Otvoriť záznam dôkazov

Výskumná otázka

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

Metóda

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.

Čo dôkazy podporujú

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

Limity a prenosnosť

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.

Klasifikácia

  • Obmedzené
  • Historické
  • Len akademické

Vplyv na NAVINES: Laboratórium pripravenosti na citácie AI

arXiv · 2024-07-02

Adversarial Search Engine Optimization for Large Language Models

Otvoriť záznam dôkazov

Výskumná otázka

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

Metóda

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

Čo dôkazy podporujú

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

Limity a prenosnosť

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.

Klasifikácia

  • Podporené
  • Obmedzené
  • Len akademické

Vplyv na NAVINES: Kontrola bezpečnosti manipulácie AI vyhľadávania

Dôkazy skontrolované 14. augusta 2026 · Neurčuje aktuálne faktory Google ani zaručený výsledok.

Výskum

Pôvodné analýzy s jasnou metódou, údajmi a obmedzeniami.

Nenaháňame algoritmy. Budujeme zrozumiteľnosť, dôveru a viditeľnosť, ktorá sa časom násobí.

Sprístupniť pokročilú inteligenciu vyhľadávania, urobiť ju užitočnou a zrozumiteľnou a pomôcť firmám premeniť ju na merateľné kroky.

Metóda

  • Určte rozsah, dôležité šablóny a ukazovatele úspechu.
  • Skontrolujte vzorku odpovedí, robots, sitemap, canonical a interné odkazy.
  • Definujte hlavnú a ochranné metriky spojené s cieľom.
  • Vytvorte porovnateľné obdobia a pohľady podľa stránky, dopytu, krajiny a zariadenia.

Obmedzenia

  • Atribúcia, súkromie a prahy údajov skrývajú časť cesty.
  • Súbežná zmena nedokazuje príčinný účinok SEO zásahu.