SoulSeek: Exploring the Use of Social Cues in LLM-based Information Seeking

TL;DR

This work investigates how integrating social cues into LLM-based information seeking on social-content platforms affects user perception, behavior, and reflection. Through design workshops and the SoulSeek prototype, the authors study how cue-aware retrieval–generation influences outcomes and processes, demonstrating improvements in perceived usefulness, trust, control, direction, serendipity, and willingness to use. They also identify limitations in social knowledge representation and cue balancing, emphasizing the need for model-level social-context understanding and system-level transparency. The findings offer design implications for personalized cue configuration, explainable cue matching, and co-search interfaces that empower user agency and collaborative human–AI information seeking. Overall, the work advances our understanding of making LLM-based search more socially aware, controllable, and reflective of human information needs.

Abstract

Social cues, which convey others' presence, behaviors, or identities, play a crucial role in human information seeking by helping individuals judge relevance and trustworthiness. However, existing LLM-based search systems primarily rely on semantic features, creating a misalignment with the socialized cognition underlying natural information seeking. To address this gap, we explore how the integration of social cues into LLM-based search influences users' perceptions, experiences, and behaviors. Focusing on social media platforms that are beginning to adopt LLM-based search, we integrate design workshops, the implementation of the prototype system (SoulSeek), a between-subjects study, and mixed-method analyses to examine both outcome- and process-level findings. The workshop informs the prototype's cue-integrated design. The study shows that social cues improve perceived outcomes and experiences, promote reflective information behaviors, and reveal limits of current LLM-based search. We propose design implications emphasizing better social-knowledge understanding, personalized cue settings, and controllable interactions.

Figures (6)

• Figure 1: The core pipeline of an LLM-based search system.
• Figure 2: Core workflow design for integrating social cues into LLM-based search systems.
• Figure 3: Pre-search user interface in SoulSeek.
• Figure 4: In-search user interface in SoulSeek.
• Figure 5: Post-search user interface in SoulSeek.