Detecting Generated Native Ads in Conversational Search

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Schmidt, Sebastian, Zelch, Ines, Bevendorff, Janek, Stein, Benno, Hagen, Matthias, Potthast, Martin
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917653624389632
author Schmidt, Sebastian
Zelch, Ines
Bevendorff, Janek
Stein, Benno
Hagen, Matthias
Potthast, Martin
author_facet Schmidt, Sebastian
Zelch, Ines
Bevendorff, Janek
Stein, Benno
Hagen, Matthias
Potthast, Martin
contents Conversational search engines such as YouChat and Microsoft Copilot use large language models (LLMs) to generate responses to queries. It is only a small step to also let the same technology insert ads within the generated responses - instead of separately placing ads next to a response. Inserted ads would be reminiscent of native advertising and product placement, both of which are very effective forms of subtle and manipulative advertising. Considering the high computational costs associated with LLMs, for which providers need to develop sustainable business models, users of conversational search engines may very well be confronted with generated native ads in the near future. In this paper, we thus take a first step to investigate whether LLMs can also be used as a countermeasure, i.e., to block generated native ads. We compile the Webis Generated Native Ads 2024 dataset of queries and generated responses with automatically inserted ads, and evaluate whether LLMs or fine-tuned sentence transformers can detect the ads. In our experiments, the investigated LLMs struggle with the task but sentence transformers achieve precision and recall values above 0.9.
format Preprint
id arxiv_https___arxiv_org_abs_2402_04889
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Detecting Generated Native Ads in Conversational Search
Schmidt, Sebastian
Zelch, Ines
Bevendorff, Janek
Stein, Benno
Hagen, Matthias
Potthast, Martin
Information Retrieval
Computation and Language
Conversational search engines such as YouChat and Microsoft Copilot use large language models (LLMs) to generate responses to queries. It is only a small step to also let the same technology insert ads within the generated responses - instead of separately placing ads next to a response. Inserted ads would be reminiscent of native advertising and product placement, both of which are very effective forms of subtle and manipulative advertising. Considering the high computational costs associated with LLMs, for which providers need to develop sustainable business models, users of conversational search engines may very well be confronted with generated native ads in the near future. In this paper, we thus take a first step to investigate whether LLMs can also be used as a countermeasure, i.e., to block generated native ads. We compile the Webis Generated Native Ads 2024 dataset of queries and generated responses with automatically inserted ads, and evaluate whether LLMs or fine-tuned sentence transformers can detect the ads. In our experiments, the investigated LLMs struggle with the task but sentence transformers achieve precision and recall values above 0.9.
title Detecting Generated Native Ads in Conversational Search
topic Information Retrieval
Computation and Language
url https://arxiv.org/abs/2402.04889