Covariate-dependent Joint Modeling of Multivariate Ordinal Preferences and Its Connections with Comparison Models
Abstract
Multivariate ordinal data along with covariates are commonly collected in problems ranging from alignment of language models with human preferences, as well as in recommender systems. For example, data sets such as MovieLens contain several movies rated on a scale 1--5 by human users, along with their demographic information such as age or gender. Similarly, data sets such as HelpSteer collect human feedback on several attributes such as "helpfulness" or "verbosity" of LLM response on an ordinal scale, with covariates depending on the LLM prompt--response pairs. Unfortunately, the standard approaches for modeling these data (a) look at the attributes individually rather than jointly, and (b) often convert the data into pairwise or list-wise win--loss comparisons for fitting models such as Bradley--Terry and Plackett--Luce. Both of these lead to a coarsening of what is actually observed, which we address via a joint covariate-dependent consecutive ratio Markov random field model. We also show pairwise or listwise comparison models are obtained under restrictions of our joint model, and that joint modeling improves comparisons. We also develop a maximum likelihood inference procedure even in the presence of an intractable normalizer.
Related
- Bayesian Soft-Tree Poisson Process Model for Covariate-Dependent Point Patterns
- Doubly Robust Estimation under Covariate Dependent Censoring in Semi-Competing Risks Data
- Covariate-dependent Nonparametric g-modeling for regression via infinite Mixture-of-Expertizing class
- Reconciling Interpretability with Covariate-Dependent Shape Flexibility in Penalized Transformation Models for Distributional Regression
- MIMO and Multi-Group Comparison Problem in Process Control: A Multivariate Statistical Framework for Systems Represented with Frequency Response Functions
BibTeX
@misc{chen2026covariatedependent,
title = {{Covariate-dependent Joint Modeling of Multivariate Ordinal Preferences and Its Connections with Comparison Models}},
author = {Yujie Chen and Antik Chakraborty and Anindya Bhadra},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.09070},
eprint = {2610.09070},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.09070},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Covariate-dependent Joint Modeling of Multivariate Ordinal Preferences and Its Connections with Comparison Models
AU - Yujie Chen
AU - Antik Chakraborty
AU - Anindya Bhadra
PY - 2026
DA - 2026/10/06
PB - arXiv
DO - 10.48550/arxiv.2610.09070
UR - https://arxiv.org/abs/2610.09070
AB - Multivariate ordinal data along with covariates are commonly collected in problems ranging from alignment of language models with human preferences, as well as in recommender systems. For example, data sets such as MovieLens contain several movies rated on a scale 1--5 by human users, along with their demographic information such as age or gender. Similarly, data sets such as HelpSteer collect human feedback on several attributes such as "helpfulness" or "verbosity" of LLM response on an ordinal scale, with covariates depending on the LLM prompt--response pairs. Unfortunately, the standard approaches for modeling these data (a) look at the attributes individually rather than jointly, and (b) often convert the data into pairwise or list-wise win--loss comparisons for fitting models such as Bradley--Terry and Plackett--Luce. Both of these lead to a coarsening of what is actually observed, which we address via a joint covariate-dependent consecutive ratio Markov random field model. We also show pairwise or listwise comparison models are obtained under restrictions of our joint model, and that joint modeling improves comparisons. We also develop a maximum likelihood inference procedure even in the presence of an intractable normalizer.
ER -