TRACE: Tourism Recommendation with Accountable Citation Evidence
Not on the map. Why?
Abstract (arXiv)
Tourism is a high-stakes setting for conversational recommender systems (CRS): a plausible-sounding suggestion can waste real money and trip time once a traveler acts on it. Existing CRS benchmarks primarily evaluate systems with a single Recall@k score over entity mentions, and tourism-specific resources add spatial or knowledge-graph context, yet none of them couple multi-turn recommendation with verbatim review-span evidence and rejection recovery. This leaves an evaluation gap for tourism recommendation that is simultaneously trustworthy, verifiable, and adaptive: recommend the right point of interest (POI) for multi-aspect preferences (such as cuisine, price, atmosphere, walking distance), justify each suggestion with verifiable evidence from prior visitors so the traveler can act without trial and error, and recover when the first recommendation is rejected mid-dialogue. We introduce TRACE, where each item is a multi-turn tourism recommendation dialogue with review-span citations and explicit rejection turns: 10,000 dialogues over 2,400 Yelp POIs and 34,208 reviews across eight U.S. cities, paired with 14 retrieval, planning, and LLM baselines, along with 25 metrics organized under Accuracy, Grounding, and Recovery. Across these baselines, TRACE reveals the Three-Competency Gap: LLM Zero-Shot leads in closed-set Recall@1 and rejection recovery but cites less densely than retrievers; non-LLM retrievers achieve surface-verbatim grounding but with low accuracy; Multi-Review Synthesis fails at recovery. The Grounding Score agrees with human citation precision (Spearman rho=+0.80, p<10^-20), and paired t-tests reproduce the per-baseline ranking (p<0.01 on the dominant contrasts). TRACE reframes accountable tourism recommendation as a joint target (right POI, verifiable evidence, adaptive repair) rather than a single-axis leaderboard.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 46 of its 60 references that are in the map. First posted on arXiv 8 May 2026.
Builds on
46 of 60 references.
References · 46
- 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
- 2018Datasheets for Datasetscited by 1,371
- 2023DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelinescited by 803
- 2023Enabling Large Language Models to Generate Text with Citationscited by 584
- 2022ASQA: Factoid Questions Meet Long-Form Answerscited by 231
- 2018Towards Deep Conversational Recommendationscited by 174
- 2020A Survey on Conversational Recommender Systemscited by 142
- 2022FaithDial: A Faithful Benchmark for Information-Seeking Dialoguecited by 108
- 2020Towards Conversational Recommendation over Multi-Type Dialogscited by 107
- 2020Towards Topic-Guided Conversational Recommender Systemcited by 94
- 2020INSPIRED: Toward Sociable Recommendation Dialog Systemscited by 91
- 2022Towards Unified Conversational Recommender Systems via Knowledge-Enhanced Prompt Learningcited by 85
34 more
- 2017Joint Deep Modeling of Users and Items Using Reviews for Recommendationcited by 84
- 2024A Review of Modern Recommender Systems Using Generative Models (Gen-RecSys)cited by 58
- 2025The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Inputcited by 49
- 2021DuRecDial 2.0: A Bilingual Parallel Corpus for Conversational Recommendationcited by 46
- 2024ITINERA: Integrating Spatial Optimization with Large Language Models for Open-domain Urban Itinerary Planningcited by 33
- 2021Improving Conversational Recommendation Systems' Quality with Context-Aware Item Meta Informationcited by 26
- 2023Citation: A Key to Building Responsible and Accountable Large Language Modelscited by 20
- 2024Retrieval-Augmented Conversational Recommendation with Prompt-based Semi-Structured Natural Language State Trackingcited by 18
- 2024TRACE the Evidence: Constructing Knowledge-Grounded Reasoning Chains for Retrieval-Augmented Generationcited by 16
- 2023TREA: Tree-Structure Reasoning Schema for Conversational Recommendationcited by 15
- 2024Pearl: A Review-driven Persona-Knowledge Grounded Conversational Recommendation Datasetcited by 15
- 2024ReasoningRec: Bridging Personalized Recommendations and Human-Interpretable Explanations through LLM Reasoningcited by 14
- 2025Spatial-RAG: Spatial Retrieval Augmented Generation for Real-World Geospatial Reasoning Questionscited by 13
- 2025TP-RAG: Benchmarking Retrieval-Augmented Large Language Model Agents for Spatiotemporal-Aware Travel Planningcited by 12
- 2025RETAIL: Towards Real-world Travel Planning for Large Language Modelscited by 9
- 2023Explicit Knowledge Graph Reasoning for Conversational Recommendationcited by 8
- 2025SynthTRIPs: A Knowledge-Grounded Framework for Benchmark Query Generation for Personalized Tourism Recommenderscited by 8
- 2026Fin-RATE: A Real-world Financial Analytics and Tracking Evaluation Benchmark for LLMs on SEC Filingscited by 8
- 2024Item-Language Model for Conversational Recommendationcited by 7
- 2024CTourLLM: Enhancing LLMs with Chinese Tourism Knowledgecited by 7
- 2025Cite Before You Speak: Enhancing Context-Response Grounding in E-commerce Conversational LLM-Agentscited by 7
- 2025Collab-REC: An LLM-based Agentic Framework for Balancing Recommendations in Tourismcited by 7
- 2025Toward Faithful Retrieval-Augmented Generation with Sparse Autoencoderscited by 6
- 2024Bridging Conversational and Collaborative Signals for Conversational Recommendationcited by 5
- 2026HalluGuard: Demystifying Data-Driven and Reasoning-Driven Hallucinations in LLMscited by 5
- 2025Should We Tailor the Talk? Understanding the Impact of Conversational Styles on Preference Elicitation in Conversational Recommender Systemscited by 4
- 2025VOGUE: A Multimodal Dataset for Conversational Recommendation in Fashioncited by 4
- 2017PERS: A Personalized and Explainable POI Recommender Systemcited by 2
- 2025Retrieval-Augmented Recommendation Explanation Generation with Hierarchical Aggregationcited by 2
- 2019Recommendation Systems for Tourism Based on Social Networks: A Surveycited by 1
- 2025Spatially-Enhanced Retrieval-Augmented Generation for Walkability and Urban Discoverycited by 1
- 2026Recommendation-as-Experience: A framework for context-sensitive adaptation in conversational recommender systemscited by 1
- 2026Reasoning-guided Collaborative Filtering with Language Models for Explainable Recommendationcited by 1
- 2026TRACE: A Conversational Framework for Sustainable Tourism Recommendation with Agentic Counterfactual Explanationscited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.