Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces Autonomous Social-Aware Relevance Learning (ASARL), a novel framework designed to enhance relevance in social search by addressing challenges such as contextual discrepancies and data scarcity. ASARL employs a multi-agent system for data curation, where ReasonAgent generates relevance labels based on social attributes, CriticAgent ensures logical consistency, and GenAgent creates synthetic query-title pairs to enrich the dataset. The framework's three-stage training process significantly improves relevance metrics and user engagement on the QQ search platform, demonstrating the effectiveness of socially aware data governance combined with preference-aligned training.
ASARL achieves remarkable gains in search relevance and user engagement by leveraging a multi-agent system that understands social context and user behavior.
The rapid growth of online social platforms has transformed communication and information retrieval, giving rise to social search, where queries-titles are typically expressed in informal, community-specific language. While large language models provide strong general-purpose semantic understanding, their effectiveness in social search is constrained by contextual discrepancy, data scarcity, and behavior-driven dynamics. To address these challenges, we propose the Autonomous Social-Aware Relevance Learning (ASARL), a fully automated framework that integrates multi-agent data curation with staged model training. ASARL leverages a collaborative agent system: ReasonAgent generates interpretable relevance labels grounded in social attributes, CriticAgent validates and ensures logical consistency, and GenAgent augments long-tail data through synthetic query-title pairs. Building on the curated dataset, ASARL employs three-stage training: Social Context Training (SCT) to capture social language patterns, Preference-Guided Optimization (PGO) to align model predictions with behavioral signals, and Social Distillation (SD) to transfer these improvements into compact models for efficient deployment. Extensive offline and online experiments on the QQ search platform demonstrate significant improvements in both offline relevance metrics and online user engagement indicators, along with enhanced annotation efficiency. These results validate the effectiveness of combining autonomous, socially grounded data governance with preference-aligned training in practical search systems.