GMTRouter: Personalized LLM Router over Multi-turn User Interactions

ACL ARR 2026 January Submission8983 Authors

Keywords: LLM routing, Preference Learning, Heterogeneous GNN

Abstract: Large Language Model (LLM) routing has demonstrated strong capability in balancing response quality with computational cost. As users exhibit diverse preferences, personalization has attracted increasing attention in LLM routing, since even identical queries may require different models to generate responses tailored to individual needs. However, existing approaches are not fully personalized and often fail to faithfully capture the complex interactions between users and LLMs. Moreover, user preference data is typically scarce and inconsistent in format, which limits the effectiveness of methods that directly leverage user-specific data. To address these challenges, we propose $\textit{GMTRouter}$, which represents multi-turn user–LLM interactions as a heterogeneous graph with five node types: user, LLM, query, response and turn, thereby maximally preserving the rich relational structure of the interaction. Through a lightweight inductive graph learning framework combined with a tailored user-conditioned graph sampling mechanism, $\textit{GMTRouter}$ learns to capture user preferences from few-shot data, enabling effective personalization. Extensive experiments demonstrate that $\textit{GMTRouter}$ outperforms the strongest baselines, achieving up to a 0.105 absolute improvement in accuracy and a 0.12 improvement in AUC. More importantly, we further demonstrate that $\textit{GMTRouter}$ can adapt to new users using only few-shot data, without extensive fine-tuning.

Paper Type: Long

Research Area: NLP Applications

Research Area Keywords: graph-based methods, generalization, conversational modeling

Contribution Types: NLP engineering experiment, Approaches to low-resource settings, Approaches low compute settings-efficiency, Data analysis

Languages Studied: English

Submission Number: 8983