This paper proposes Popularity-Aware Denoising (PAD), a framework to improve denoising recommendation methods by modulating denoising strength based on item popularity.

The paper introduces SB-ECC, a novel score-based decoder that models error correction as continuous-time denoising, achieving state-of-the-art performance across various code families and noise levels…

This paper proposes a two-stage algorithm, AC-IHT, for high-dimensional regression with contamination, achieving near-optimal estimation and strong oracle property.

The paper introduces Score Broadcast and Decorrelation (SBD), a general theoretical framework that unifies broadcast-based credit assignment across various differentiable loss functions by leveraging…

Proposed DDMSR framework for multi-modal sequential recommendation using graph-based feature denoising and frequency-domain sequence denoising, and multi-modal contrastive alignment objective.

This paper establishes a connection between neural network approximation of score functions and approximation of probability distributions generated by reverse diffusion models.

The paper proposes Alignment-Guided Score Matching (AGSM), a lightweight, reward-free post-training method that integrates contrastive alignment guidance directly into the score-matching objective of…

The paper identifies a fundamental mismatch between standard pairwise ranking metrics (like AP and FPR-95) and the true assignment objective in multi-view object association, proposing a Sinkhorn-base…

This paper identifies prediction bias, a failure mode of entropy minimization in test-time adaptation, and proposes Distribution Shift Bias Reduction (DSBR) to stabilize adaptation and prevent model c…

The paper introduces and analyzes several novel data appraisal metrics, including the Vendi Score and matrix spectral functions, demonstrating that efficient optimization techniques make these metrics…

This paper introduces CODA, a real-time score following system that exploits the cascaded structure of music scores for prediction consistency and enables recovery from score discontinuities.

This paper shows that small forward-marginal error in score matching does not guarantee numerical stability, and constructs a smooth score field with small forward-marginal error but diverging moments…

This paper derives the Oracle Distance theorem for discrete diffusion models and proves that the negative ELBO is equal to the data entropy plus the path KL from the oracle reverse process to the lear…

The paper introduces DSL-LLaDA, a method that lightly adapts a pre-trained masked diffusion language model to perform continuous denoising in embedding space, significantly improving text generation q…

This paper proposes a two-stage MSR system, DTT-BSR+, that decouples distribution fitting from signal reconstruction for improved accuracy and semantic consistency.

The paper repurposes a pre-trained speech classifier as the backbone for diffusion generation, reducing the need for two separately trained models.

The paper describes how to compute singular value soft-thresholding using matrix polar decomposition for faster GPU processing.

This paper models the changes in graph adjacency matrices over time as a low-rank matrix and develops a method for jointly interpolating signals and estimating graph adjacency matrices using this assu…

This paper introduces Spectral Attention and Graph Convolutional Attention (GCA) for denoising graphs, which outperforms linear attention and provably utilizes the input graph spectrum.