Research map / Large language models / Retrieval and knowledge
Enhancing Conversational Recommender Systems with Tree-Structured Knowledge and Pretrained Language Models
Yongwen Ren, Chao Wang, Peng Du, Chuan Qin, Dazhong Shen, Hui Xiong
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add PCRS-TKA modules on a BERT encoder (ablation) | DCRS-style baseline with BERT · INSPIRED | mrr@10: 0.138 -> 0.148 | Improved |
| add PCRS-TKA modules on a BERT encoder (ablation) | baseline with BERT · INSPIRED | recall@10: 0.250 -> 0.254 | Improved |
| add PCRS-TKA modules on a BERT encoder (ablation) | baseline with BERT · INSPIRED | ndcg@10: 0.162 -> 0.174 | Improved |
| add PCRS-TKA modules on a GPT-2 generator (ablation) | baseline with GPT-2 · INSPIRED | recall@10: 0.230 -> 0.258 | Improved |
| add PCRS-TKA modules on a GPT-2 generator (ablation) | baseline with GPT-2 · INSPIRED | mrr@10: 0.129 -> 0.140 | Improved |
| add PCRS-TKA modules on original RoBERTa/DialoGPT (ablation) | UniCRS-style baseline · INSPIRED | mrr@10: 0.131 -> 0.158 | Improved |
| add PCRS-TKA modules on original RoBERTa/DialoGPT (ablation) | UniCRS-style baseline · INSPIRED | recall@50: 0.406 -> 0.438 | Improved |
| add knowledge-tree prompts, user preference prompts, multimodal integration (each) (ablation) | PCRS-TKA · INSPIRED | recommendation metrics | Improved |
| use hierarchical tree instead of unordered triples (ablation) | PCRS-TKA · INSPIRED | recommendation metrics | Improved |
| add context-aware knowledge filtering (ablation) | PCRS-TKA · INSPIRED | recommendation metrics | Improved |
| increase knowledge tree degree (two-layer trees) | PCRS-TKA · INSPIRED | recommendation metrics | Hurt |
| use two-layer instead of one-layer knowledge tree | PCRS-TKA · INSPIRED | recommendation metrics | No clear effect |
Closest papers by meaning
- Bridging the User-side Knowledge Gap in Knowledge-aware Recommendations with Large Language Models (AAAI 2025 · similarity 0.89)
- ConvRec-R1: Training LLM-based Conversational Recommender Systems with Reinforcement Learning (ICLR 2026 · similarity 0.88)
- A Goal Interaction Graph Planning Framework for Conversational Recommendation (AAAI 2024 · similarity 0.88)
- Semantic Convergence: Harmonizing Recommender Systems via Two-Stage Alignment and Behavioral Semantic Tokenization (AAAI 2025 · similarity 0.88)
- LLMRG: Improving Recommendations through Large Language Model Reasoning Graphs (AAAI 2024 · similarity 0.87)
- ESPRESSO: An Effective Approach to Passage Retrieval for High-Quality Conversational Recommender Systems (AAAI 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
