RAG-Enhanced Large Language Models for Dynamic Content Expiration Prediction in Web Search

TL;DR

This paper introduces a novel LLM-based framework for predicting content expiration in web search, improving freshness and user satisfaction by understanding semantic validity tailored to user queries.

Contribution

It presents a new dynamic validity inference approach using LLMs to determine content expiration, addressing limitations of static filtering methods in search engines.

Findings

Significant improvements in search freshness metrics.

Enhanced user experience through better content relevance.

Effective LLM reasoning for semantic expiration detection.

Abstract

In commercial web search, aligning content freshness with user intent remains challenging due to the highly varied lifespans of information. Traditional industrial approaches rely on static time-window filtering, resulting in "one-size-fits-all" rankings where content may be chronologically recent but semantically expired. To address the limitation, we present a novel Large Language Models (LLMs)-based Query-Aware Dynamic Content Expiration Prediction Framework deployed in Baidu search, reformulating timeliness as a dynamic validity inference task. Our framework extracts fine-grained temporal contexts from documents and leverages LLMs to deduce a query-specific "validity horizon"-a semantic boundary defining when information becomes obsolete based on user intent. Integrated with robust hallucination mitigation strategies to ensure reliability, our approach has been evaluated through offline and online A/B testing on live production traffic. Results demonstrate significant improvements in search freshness and user experience metrics, validating the effectiveness of LLM-driven reasoning for solving semantic expiration at an industrial scale.