RankList -- A Listwise Preference Learning Framework for Predicting Subjective Preferences

TL;DR

RankList introduces a listwise preference learning framework that models global ranking constraints, improving subjective preference prediction in tasks like speech emotion recognition and image aesthetics over traditional pairwise methods.

Contribution

The paper proposes a novel listwise preference learning framework, RankList, that generalizes RankNet with list-level supervision and skip-wise comparisons for better global ranking accuracy.

Findings

Outperforms baseline methods on benchmark SER datasets in Kendall's Tau and accuracy.

Demonstrates broad applicability across different modalities like speech and images.

Shows improved cross-domain generalization and ranking fidelity.

Abstract

Preference learning has gained significant attention in tasks involving subjective human judgments, such as \emph{speech emotion recognition} (SER) and image aesthetic assessment. While pairwise frameworks such as RankNet offer robust modeling of relative preferences, they are inherently limited to local comparisons and struggle to capture global ranking consistency. To address these limitations, we propose RankList, a novel listwise preference learning framework that generalizes RankNet to structured list-level supervision. Our formulation explicitly models local and non-local ranking constraints within a probabilistic framework. The paper introduces a log-sum-exp approximation to improve training efficiency. We further extend RankList with skip-wise comparisons, enabling progressive exposure to complex list structures and enhancing global ranking fidelity. Extensive experiments demonstrate the superiority of our method across diverse modalities. On benchmark SER datasets (MSP-Podcast, IEMOCAP, BIIC Podcast), RankList achieves consistent improvements in Kendall's Tau and ranking accuracy compared to standard listwise baselines. We also validate our approach on aesthetic image ranking using the Artistic Image Aesthetics dataset, highlighting its broad applicability. Through ablation and cross-domain studies, we show that RankList not only improves in-domain ranking but also generalizes better across datasets. Our framework offers a unified, extensible approach for modeling ordered preferences in subjective learning scenarios.