Where and What: Reasoning Dynamic and Implicit Preferences in Situated Conversational Recommendation

TL;DR

This paper introduces SiPeR, a framework for reasoning about dynamic and implicit user preferences in situated conversational recommendation, improving accuracy and response quality by integrating scene transition estimation and Bayesian inference.

Contribution

The paper presents a novel framework, SiPeR, that enhances situated conversational recommendation by explicitly modeling scene satisfaction and user preferences using multimodal large language models.

Findings

SiPeR outperforms baselines in recommendation accuracy.

SiPeR improves response generation quality.

Experiments validate the effectiveness of scene transition estimation and Bayesian inference.

Abstract

Situated conversational recommendation (SCR), which utilizes visual scenes grounded in specific environments and natural language dialogue to deliver contextually appropriate recommendations, has emerged as a promising research direction due to its close alignment with real-world scenarios. Compared to traditional recommendations, SCR requires a deeper understanding of dynamic and implicit user preferences, as the surrounding scene often influences users' underlying interests, while both may evolve across conversations. This complexity significantly impacts the timing and relevance of recommendations. To address this, we propose situated preference reasoning (SiPeR), a novel framework that integrates two core mechanisms: (1) Scene transition estimation, which estimates whether the current scene satisfies user needs, and guides the user toward a more suitable scene when necessary; and (2) Bayesian inverse inference, which leverages the likelihood of multimodal large language models (MLLMs) to predict user preferences about candidate items within the scene. Extensive experiments on two representative benchmarks demonstrate SiPeR's superiority in both recommendation accuracy and response generation quality. The code and data are available at https://github.com/DongdingLin/SiPeR.