Content Moderation in TV Search: Balancing Policy Compliance, Relevance, and User Experience
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908376115445760 |
|---|---|
| author | Hande, Adeep Sundararajan, Kishorekumar Hamidian, Sardar Ture, Ferhan |
| author_facet | Hande, Adeep Sundararajan, Kishorekumar Hamidian, Sardar Ture, Ferhan |
| contents | Millions of people rely on search functionality to find and explore content on entertainment platforms. Modern search systems use a combination of candidate generation and ranking approaches, with advanced methods leveraging deep learning and LLM-based techniques to retrieve, generate, and categorize search results. Despite these advancements, search algorithms can still surface inappropriate or irrelevant content due to factors like model unpredictability, metadata errors, or overlooked design flaws. Such issues can misalign with product goals and user expectations, potentially harming user trust and business outcomes. In this work, we introduce an additional monitoring layer using Large Language Models (LLMs) to enhance content moderation. This additional layer flags content if the user did not intend to search for it. This approach serves as a baseline for product quality assurance, with collected feedback used to refine the initial retrieval mechanisms of the search model, ensuring a safer and more reliable user experience. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_17207 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Content Moderation in TV Search: Balancing Policy Compliance, Relevance, and User Experience Hande, Adeep Sundararajan, Kishorekumar Hamidian, Sardar Ture, Ferhan Information Retrieval Machine Learning Millions of people rely on search functionality to find and explore content on entertainment platforms. Modern search systems use a combination of candidate generation and ranking approaches, with advanced methods leveraging deep learning and LLM-based techniques to retrieve, generate, and categorize search results. Despite these advancements, search algorithms can still surface inappropriate or irrelevant content due to factors like model unpredictability, metadata errors, or overlooked design flaws. Such issues can misalign with product goals and user expectations, potentially harming user trust and business outcomes. In this work, we introduce an additional monitoring layer using Large Language Models (LLMs) to enhance content moderation. This additional layer flags content if the user did not intend to search for it. This approach serves as a baseline for product quality assurance, with collected feedback used to refine the initial retrieval mechanisms of the search model, ensuring a safer and more reliable user experience. |
| title | Content Moderation in TV Search: Balancing Policy Compliance, Relevance, and User Experience |
| topic | Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2505.17207 |