From Noise to Order: Learning to Rank via Denoising Diffusion

2026-02-12

AI Summary

This work proposes DiffusionRank, a novel approach that introduces denoising diffusion generative models to learning-to-rank for the first time. Traditional ranking methods predominantly rely on discriminative models, which struggle to capture the full joint distribution between query-document features and relevance labels, thereby limiting model robustness. In contrast, DiffusionRank explicitly models this joint distribution and reformulates pointwise and pairwise ranking objectives in a generative framework. Built upon the TabDiff architecture, the proposed method achieves significant performance gains over existing discriminative models across multiple ranking benchmarks, demonstrating the effectiveness and potential of generative modeling in information retrieval.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

In information retrieval (IR), learning-to-rank (LTR) methods have traditionally limited themselves to discriminative machine learning approaches that model the probability of the document being relevant to the query given some feature representation of the query-document pair. In this work, we propose an alternative denoising diffusion-based deep generative approach to LTR that instead models the full joint distribution over feature vectors and relevance labels. While in the discriminative setting, an over-parameterized ranking model may find different ways to fit the training data, we hypothesize that candidate solutions that can explain the full data distribution under the generative setting produce more robust ranking models. With this motivation, we propose DiffusionRank that extends TabDiff, an existing denoising diffusion-based generative model for tabular datasets, to create generative equivalents of classical discriminative pointwise and pairwise LTR objectives. Our empirical results demonstrate significant improvements from DiffusionRank models over their discriminative counterparts. Our work points to a rich space for future research exploration on how we can leverage ongoing advancements in deep generative modeling approaches, such as diffusion, for learning-to-rank in IR.

Problem

• learning-to-rank
• information retrieval
• denoising diffusion
• generative modeling
• robust ranking

Innovation

• denoising diffusion
• learning-to-rank
• generative modeling
• DiffusionRank
• information retrieval