Summary of Zero-shot Image Moderation in Google Ads with Llm-assisted Textual Descriptions and Cross-modal Co-embeddings, by Enming Luo et al.
Zero-Shot Image Moderation in Google Ads with LLM-Assisted Textual Descriptions and Cross-modal Co-embeddings
by Enming Luo, Wei Qiao, Katie Warren, Jingxiang Li, Eric Xiao, Krishna Viswanathan, Yuan Wang, Yintao Liu, Jimin Li, Ariel Fuxman
First submitted to arxiv on: 18 Dec 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed method utilizes human-curated textual descriptions and cross-modal text-image co-embeddings to enable zero-shot classification of policy-violating ads images at Google, addressing massive volumes of diverse ads with evolving policies. By leveraging large language models (LLMs) and user expertise, the system generates and refines a comprehensive set of textual descriptions representing policy guidelines. During inference, co-embedding similarity between incoming images and the textual descriptions serves as a reliable signal for policy violation detection, enabling efficient and adaptable ads content moderation. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper presents an innovative way to quickly identify ads that break Google’s rules without needing a lot of labeled data or human help. It uses special word embeddings to connect text and image features, so it can recognize problematic ads even if they’re never seen before. This approach could be very useful for Google’s massive ad moderation task. |
Keywords
» Artificial intelligence » Classification » Embedding » Inference » Zero shot
