Multi-Domain Explainability of Preferences

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Calderon, Nitay, Ein-Dor, Liat, Reichart, Roi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918037970485248
author Calderon, Nitay
Ein-Dor, Liat
Reichart, Roi
author_facet Calderon, Nitay
Ein-Dor, Liat
Reichart, Roi
contents Preference mechanisms, such as human preference, LLM-as-a-Judge (LaaJ), and reward models, are central to aligning and evaluating large language models (LLMs). Yet, the underlying concepts that drive these preferences remain poorly understood. In this work, we propose a fully automated method for generating local and global concept-based explanations of preferences across multiple domains. Our method utilizes an LLM to identify concepts that distinguish between chosen and rejected responses, and to represent them with concept-based vectors. To model the relationships between concepts and preferences, we propose a white-box Hierarchical Multi-Domain Regression model that captures both domain-general and domain-specific effects. To evaluate our method, we curate a dataset spanning eight challenging and diverse domains and explain twelve mechanisms. Our method achieves strong preference prediction performance, outperforming baselines while also being explainable. Additionally, we assess explanations in two application-driven settings. First, guiding LLM outputs with concepts from LaaJ explanations yields responses that those judges consistently prefer. Second, prompting LaaJs with concepts explaining humans improves their preference predictions. Together, our work establishes a new paradigm for explainability in the era of LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2505_20088
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Multi-Domain Explainability of Preferences
Calderon, Nitay
Ein-Dor, Liat
Reichart, Roi
Computation and Language
Preference mechanisms, such as human preference, LLM-as-a-Judge (LaaJ), and reward models, are central to aligning and evaluating large language models (LLMs). Yet, the underlying concepts that drive these preferences remain poorly understood. In this work, we propose a fully automated method for generating local and global concept-based explanations of preferences across multiple domains. Our method utilizes an LLM to identify concepts that distinguish between chosen and rejected responses, and to represent them with concept-based vectors. To model the relationships between concepts and preferences, we propose a white-box Hierarchical Multi-Domain Regression model that captures both domain-general and domain-specific effects. To evaluate our method, we curate a dataset spanning eight challenging and diverse domains and explain twelve mechanisms. Our method achieves strong preference prediction performance, outperforming baselines while also being explainable. Additionally, we assess explanations in two application-driven settings. First, guiding LLM outputs with concepts from LaaJ explanations yields responses that those judges consistently prefer. Second, prompting LaaJs with concepts explaining humans improves their preference predictions. Together, our work establishes a new paradigm for explainability in the era of LLMs.
title Multi-Domain Explainability of Preferences
topic Computation and Language
url https://arxiv.org/abs/2505.20088