Latent Preference Modeling for Cross-Session Personalized Tool Calling

TL;DR

This paper introduces MPT, a benchmark for personalized tool calling in multi-session dialogues, and proposes PRefine, a memory-augmented method that enhances tool accuracy by modeling user preferences as evolving hypotheses.

Contribution

The paper presents a new benchmark and a novel test-time memory-augmented approach for personalized tool calling in multi-session dialogues.

Findings

PRefine improves tool-calling accuracy using only 1.24% of tokens compared to full-history prompts.

MPT benchmark covers three key challenges: Preference Recall, Induction, and Transfer.

Memory capturing user preference reasons enhances personalization in agent systems.

Abstract

Users often omit essential details in their requests to LLM-based agents, resulting in under-specified inputs for tool use. This poses a fundamental challenge for tool-augmented agents, as API execution typically requires complete arguments, highlighting the need for personalized tool calling. To study this problem, we introduce MPT, a benchmark comprising 265 multi-session dialogues that cover three challenges: Preference Recall, Preference Induction, and Preference Transfer. We also propose PRefine, a test-time memory-augmented method that represents user preferences as evolving hypotheses. Through a generate--verify--refine loop, it extracts reusable constraints from history and improves tool-calling accuracy while using only 1.24% of the tokens required by full-history prompting. These results indicate that robust personalization in agentic systems depends on memory that captures the reasons behind user choices, not just the choices themselves.