Multi-Dimensional Evaluation of Sustainable City Trips with LLM-as-a-Judge and Human-in-the-Loop

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Banerjee, Ashmi, Satish, Adithi, Wörndl, Wolfgang, Deldjoo, Yashar
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917439274483712
author Banerjee, Ashmi
Satish, Adithi
Wörndl, Wolfgang
Deldjoo, Yashar
author_facet Banerjee, Ashmi
Satish, Adithi
Wörndl, Wolfgang
Deldjoo, Yashar
contents Evaluating nuanced conversational travel recommendations is challenging when human annotations are costly and standard metrics ignore stakeholder-centric goals. We study LLMs-as-Judges for sustainable city-trip lists across four dimensions -- relevance, diversity, sustainability, and popularity balance, and propose a three-phase calibration framework: (1) baseline judging with multiple LLMs, (2) expert evaluation to identify systematic misalignment, and (3) dimension-specific calibration via rules and few-shot examples. Across two recommendation settings, we observe model-specific biases and high dimension-level variance, even when judges agree on overall rankings. Calibration clarifies reasoning per dimension but exposes divergent interpretations of sustainability, highlighting the need for transparent, bias-aware LLM evaluation. Prompts and code are released for reproducibility: https://github.com/ashmibanerjee/trs-llm-calibration.
format Preprint
id arxiv_https___arxiv_org_abs_2604_24158
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Multi-Dimensional Evaluation of Sustainable City Trips with LLM-as-a-Judge and Human-in-the-Loop
Banerjee, Ashmi
Satish, Adithi
Wörndl, Wolfgang
Deldjoo, Yashar
Artificial Intelligence
Evaluating nuanced conversational travel recommendations is challenging when human annotations are costly and standard metrics ignore stakeholder-centric goals. We study LLMs-as-Judges for sustainable city-trip lists across four dimensions -- relevance, diversity, sustainability, and popularity balance, and propose a three-phase calibration framework: (1) baseline judging with multiple LLMs, (2) expert evaluation to identify systematic misalignment, and (3) dimension-specific calibration via rules and few-shot examples. Across two recommendation settings, we observe model-specific biases and high dimension-level variance, even when judges agree on overall rankings. Calibration clarifies reasoning per dimension but exposes divergent interpretations of sustainability, highlighting the need for transparent, bias-aware LLM evaluation. Prompts and code are released for reproducibility: https://github.com/ashmibanerjee/trs-llm-calibration.
title Multi-Dimensional Evaluation of Sustainable City Trips with LLM-as-a-Judge and Human-in-the-Loop
topic Artificial Intelligence
url https://arxiv.org/abs/2604.24158