Research map / Large language models / Retrieval and knowledge
Factual and Informative Review Generation for Explainable Recommendation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use personalized retrieval and a reader trained to ground answers in retrieved reviews | PRAG · Amazon Movies and TV, TripAdvisor, Yelp | Entail: 25.6 -> 88.8; 22.8 -> 80.1; 35.9 -> 86.2 | Helped |
| use personalized retrieval and keyword-guided generation | PRAG · Amazon Movies and TV, TripAdvisor, Yelp | D-1: 39.9 -> 45.6; 42.6 -> 39.9; 43.1 -> 47.1 | Mixed |
| use personalized retrieval and keyword-guided generation | PRAG · Amazon Movies and TV, TripAdvisor, Yelp | MAUVE: 12.9 -> 23.1; 5.3 -> 42.8; 10.4 -> 20.3 | Helped |
| use T5 sentence embeddings instead of MPNET (ablation) | PRAG · Amazon Movies and TV, TripAdvisor, Yelp | RMSE: 0.79 -> 0.80; 0.95 -> 0.96; 1.01 -> 1.00 | No clear effect |
| tie Q to the user factor rather than the item factor (ablation) | PRAG · Amazon Movies and TV, TripAdvisor, Yelp | RMSE: 0.79 -> 0.80; 0.95 -> 0.95; 1.01 -> 1.01 | Mixed |
| marginalize the universal query component before retrieval (ablation) | PRAG retriever · Amazon Movies and TV, TripAdvisor, Yelp | agreement-at-5: 0.54 -> 2.87; 0.61 -> 2.44; 0.27 -> 2.50 | Helped |
| condition generation on retrieved reviews rather than only user/item vectors | retriever-augmented generators · Amazon Movies and TV, TripAdvisor, Yelp | generation diversity and quality | Helped |
| use retrieved reviews with a generator trained directly on noisy review targets | Optimus and SUM · Amazon Movies and TV, TripAdvisor, Yelp | Entail | No clear effect |
Closest papers by meaning
- Coherency Improved Explainable Recommendation via Large Language Model (AAAI 2025 · similarity 0.86)
- Fine-Tuning Large Language Model Based Explainable Recommendation with Explainable Quality Reward (AAAI 2024 · similarity 0.83)
- Multimodal Quantitative Language for Generative Recommendation (ICLR 2025 · similarity 0.82)
- Echoes in Filter Bubble: Diagnosing and Curing Popularity Bias in Generative Recommender Systems (NeurIPS 2026 · similarity 0.82)
- GEAR-Up: Generative AI and External Knowledge-Based Retrieval: Upgrading Scholarly Article Searches for Systematic Reviews (AAAI 2024 · similarity 0.81)
- Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations (ICML 2024 · similarity 0.81)
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.