Lune

SIGIR2025Top-tier venue

Beyond Whole Dialogue Modeling: Contextual Disentanglement for Conversational Recommendation

Guojia An, Jie Zou, Jiwei Wei, Chaoning Zhang, Fuming Sun, Yang Yang

Abstract

Conversational recommender systems aim to provide personalized recommendations by analyzing and utilizing contextual information related to dialogue. However, existing methods typically model the dialogue context as a whole, neglecting the inherent complexity and entanglement within the dialogue. Specifically, a dialogue comprises both focus information and background information, which mutually influence each other. Current methods tend to model these two types of information mixedly, leading to misinterpretation of users' actual needs, thereby lowering the accuracy of recommendations. To address this issue, this paper proposes a novel model to introduce contextual disentanglement for improving conversational recommender systems, named DisenCRS. The proposed model DisenCRS employs a dual disentanglement framework, including self-supervised contrastive disentanglement and counterfactual inference disentanglement, to effectively distinguish focus information and background information from the dialogue context under unsupervised conditions. Moreover, we design an adaptive prompt learning module to automatically select the most suitable prompt based on the specific dialogue context, fully leveraging the power of large language models. Experimental results on two widely used public datasets demonstrate that DisenCRS significantly outperforms existing conversational recommendation models, achieving superior performance on both item recommendation and response generation tasks.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

7 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Not All Information Brings Benefits: Personalization-Driven Agent Debate for Conversational Recommendation cites MSCRS: Multi-modal Semantic Graph Prompt Learning Framework for Conversational Recommender Systems
  • Not All Information Brings Benefits: Personalization-Driven Agent Debate for Conversational Recommendation cites Unleashing the Potential of Neighbors: Diffusion-based Latent Neighbor Generation for Session-based Recommendation
  • Not All Information Brings Benefits: Personalization-Driven Agent Debate for Conversational Recommendation cites Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion
  • MSCRS: Multi-modal Semantic Graph Prompt Learning Framework for Conversational Recommender Systems cites BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
  • MSCRS: Multi-modal Semantic Graph Prompt Learning Framework for Conversational Recommender Systems cites Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion
  • MSCRS: Multi-modal Semantic Graph Prompt Learning Framework for Conversational Recommender Systems cites Interactive Path Reasoning on Graph for Conversational Recommendation
  • From Dialogue to Destination: Geography-Aware Large Language Models with Multimodal Fusion for Conversational Recommendation cites MSCRS: Multi-modal Semantic Graph Prompt Learning Framework for Conversational Recommender Systems
  • From Dialogue to Destination: Geography-Aware Large Language Models with Multimodal Fusion for Conversational Recommendation cites Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion
  • Unleashing the Potential of Neighbors: Diffusion-based Latent Neighbor Generation for Session-based Recommendation cites MSCRS: Multi-modal Semantic Graph Prompt Learning Framework for Conversational Recommender Systems
  • CARD: Non-Uniform Quantization of Visual Semantic Unit for Generative Recommendation cites MSCRS: Multi-modal Semantic Graph Prompt Learning Framework for Conversational Recommender Systems
  • CARD: Non-Uniform Quantization of Visual Semantic Unit for Generative Recommendation cites From Dialogue to Destination: Geography-Aware Large Language Models with Multimodal Fusion for Conversational Recommendation
  • CARD: Non-Uniform Quantization of Visual Semantic Unit for Generative Recommendation cites Unleashing the Potential of Neighbors: Diffusion-based Latent Neighbor Generation for Session-based Recommendation
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers7

  1. MSCRS: Multi-modal Semantic Graph Prompt Learning Framework for Conversational Recommender SystemsYibiao Wei, Jie Zou, Weikang Guo, Guoqing Wang et al.SIGIR 2025 · 11 citations
  2. From Dialogue to Destination: Geography-Aware Large Language Models with Multimodal Fusion for Conversational RecommendationYeming Li, Chenxi Liu, Jie Zou, Cheng Long et al.AAAI 2026 · 3 citations
  3. Unleashing the Potential of Neighbors: Diffusion-based Latent Neighbor Generation for Session-based RecommendationYuhan Yang, Jie Zou, Guojia An, Jiwei Wei et al.KDD 2026 · 2 citations
  4. CARD: Non-Uniform Quantization of Visual Semantic Unit for Generative RecommendationYibiao Wei, Jie Zou, Pengfei Zhang, Xiao Ao et al.SIGIR 2026 · 1 citation
  5. Not All Information Brings Benefits: Personalization-Driven Agent Debate for Conversational RecommendationPengfei Zhang, Guojia An, Jin Huang, Yuhan Yang et al.WWW 2026

Builds on24

  1. BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and ComprehensionMike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad et al.ACL 2020 · 1,224 citations
  2. Disentangling User Interest and Conformity for Recommendation with Causal EmbeddingYu Zheng, Chen Gao, Xiang Li, Xiangnan He et al.WWW 2021 · 392 citations
  3. Improving Conversational Recommender Systems via Knowledge Graph based Semantic FusionKun Zhou, Wayne Xin Zhao, Shuqing Bian, Yuanhang Zhou et al.KDD 2020 · 309 citations
  4. Interactive Path Reasoning on Graph for Conversational RecommendationWenqiang Lei, Gangyi Zhang, Xiangnan He, Yisong Miao et al.KDD 2020 · 158 citations
  5. Disentangled Contrastive Collaborative FilteringXubin Ren, Lianghao Xia, Jiashu Zhao, Dawei Yin et al.SIGIR 2023 · 154 citations

Related papers

  1. Towards Unified Conversational Recommender Systems via Knowledge-Enhanced Prompt LearningXiaolei Wang, Kun Zhou, Ji-Rong Wen, Wayne Xin ZhaoKDD 2022 · 143 citations
  2. Collaborative Retrieval for Large Language Model-based Conversational Recommender SystemsYaochen Zhu, Chao Wan, Harald Steck, Dawen Liang et al.WWW 2025 · 15 citations
  3. Dynamic Open-book Prompt for Conversational Recommender SystemXuan Ma, Tieyun Qian, Ke SunEMNLP Findings 2023 · 1 citation
  4. LatentCRS: A Variational EM Framework for Bridging Semantics and Behavior in LLM-based Conversational RecommendationGuanrong Li, Kuo Tian, Jinnan Qi, Qinghan Fu et al.KDD 2026 · 1 citation
  5. User-Centric Conversational Recommendation with Multi-Aspect User ModelingShuokai Li, Ruobing Xie, Yongchun Zhu, Xiang Ao et al.SIGIR 2022 · 60 citations
Dusk over the sea between two cliffs drawn in fine vertical lines