Advertiser Content Understanding via LLMs for Google Ads Safety

• URL: http://arxiv.org/abs/2409.15343v1
• Date: Tue, 10 Sep 2024 00:57:51 GMT
• Title: Advertiser Content Understanding via LLMs for Google Ads Safety
• Authors: Joseph Wallace, Tushar Dogra, Wei Qiao, Yuan Wang,
• Abstract summary: This study proposes a method to understand advertiser's intent for content policy violations, using Large Language Models (LLMs) We generate advertiser's content profile based on multiple signals from their ads, domains, targeting info, etc. After minimal prompt tuning our method was able to reach 95% accuracy on a small test set.
• Score: 9.376815457907195
• License:
• Abstract: Ads Content Safety at Google requires classifying billions of ads for Google Ads content policies. Consistent and accurate policy enforcement is important for advertiser experience and user safety and it is a challenging problem, so there is a lot of value for improving it for advertisers and users. Inconsistent policy enforcement causes increased policy friction and poor experience with good advertisers, and bad advertisers exploit the inconsistency by creating multiple similar ads in the hope that some will get through our defenses. This study proposes a method to understand advertiser's intent for content policy violations, using Large Language Models (LLMs). We focus on identifying good advertisers to reduce content over-flagging and improve advertiser experience, though the approach can easily be extended to classify bad advertisers too. We generate advertiser's content profile based on multiple signals from their ads, domains, targeting info, etc. We then use LLMs to classify the advertiser content profile, along with relying on any knowledge the LLM has of the advertiser, their products or brand, to understand whether they are likely to violate a certain policy or not. After minimal prompt tuning our method was able to reach 95\% accuracy on a small test set.

Related papers

• On the Use of Proxies in Political Ad Targeting [49.61009579554272]
  We show that major political advertisers circumvented mitigations by targeting proxy attributes. Our findings have crucial implications for the ongoing discussion on the regulation of political advertising.
  arXiv  Detail & Related papers  (2024-10-18T17:15:13Z)
• Why am I Still Seeing This: Measuring the Effectiveness Of Ad Controls and Explanations in AI-Mediated Ad Targeting Systems [55.02903075972816]
  We evaluate the effectiveness of Meta's "See less" ad control and the actionability of ad targeting explanations following the shift to AI-mediated targeting. We find that utilizing the "See less" ad control for the topics we study does not significantly reduce the number of ads shown by Meta on these topics. We find that the majority of ad targeting explanations for local ads made no reference to location-specific targeting criteria.
  arXiv  Detail & Related papers  (2024-08-21T18:03:11Z)
• Before Blue Birds Became X-tinct: Understanding the Effect of Regime Change on Twitter's Advertising and Compliance of Advertising Policies [0.562479170374811]
  We present the first large-scale audit of advertising on Twitter focusing on compliance with the platform's advertising policies. We find evidence of widespread noncompliance with Twitter's political and adult content advertising policies. Major brands decreased their advertising on Twitter, suggesting a negative immediate effect on the platform's advertising revenue.
  arXiv  Detail & Related papers  (2023-09-22T02:39:53Z)
• Long-Term Ad Memorability: Understanding & Generating Memorable Ads [54.23854539909078]
  There has been no large-scale study on the memorability of ads. We release the first memorability dataset, LAMBDA, consisting of 1749 participants and 2205 ads covering 276 brands.
  arXiv  Detail & Related papers  (2023-09-01T10:27:04Z)
• Discrimination through Image Selection by Job Advertisers on Facebook [79.21648699199648]
  We propose and investigate the prevalence of a new means for discrimination in job advertising. It combines both targeting and delivery -- through the disproportionate representation or exclusion of people of certain demographics in job ad images. We use the Facebook Ad Library to demonstrate the prevalence of this practice.
  arXiv  Detail & Related papers  (2023-06-13T03:43:58Z)
• Problematic Advertising and its Disparate Exposure on Facebook [15.667983888666312]
  We study Facebook and investigate key gaps in our understanding of problematic online advertising. We find that older people and minority groups are especially likely to be shown such ads. Given that 22% of problematic ads had no specific targeting from advertisers, we infer that ad delivery algorithms played a significant role in the biased distribution of these ads.
  arXiv  Detail & Related papers  (2023-06-09T17:23:59Z)
• AI-Driven Contextual Advertising: A Technology Report and Implication Analysis [0.0]
  Programmatic advertising consists in automated auctioning of digital ad space. The interest in contextual advertising is in part a counterreaction to the current dependency on personal data. Developments in Artificial Intelligence (AI) allow for a deeper semantic understanding of context.
  arXiv  Detail & Related papers  (2022-05-02T13:44:58Z)
• Auditing for Discrimination in Algorithms Delivering Job Ads [70.02478301291264]
  We develop a new methodology for black-box auditing of algorithms for discrimination in the delivery of job advertisements. Our first contribution is to identify the distinction between skew in ad delivery due to protected categories such as gender or race. Second, we develop an auditing methodology that distinguishes between skew explainable by differences in qualifications from other factors. Third, we apply our proposed methodology to two prominent targeted advertising platforms for job ads: Facebook and LinkedIn.
  arXiv  Detail & Related papers  (2021-04-09T17:38:36Z)
• An Empirical Study of In-App Advertising Issues Based on Large Scale App Review Analysis [67.58267006314415]
  We present a large-scale analysis on ad-related user feedback from App Store and Google Play. From a statistical analysis of 36,309 ad-related reviews, we find that users care most about the number of unique ads and ad display frequency during usage. Some ad issue types are addressed more quickly by developers than other ad issues.
  arXiv  Detail & Related papers  (2020-08-22T05:38:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.