LACU

[arXiv]

Locality-aware continual unlearning (LACU), a framework that stabilizes sequential concept removal

GUDA

[arXiv]

Efficiently estimate which training data groups influenced diffusion model outputs

Demystifying MaskGIT

[arXiv]

Theoretically analyzes the MaskGIT sampler, poviding a choose-then-sample (CTS) formulation

MF-RAE

[arXiv]

Training flow-map models in RAE latent with consistency mid-training for trajectory-aware initialization

ConceptTRAK

[arXiv]

A framework for Identify which training examples influenced specific concepts within the diffusion model

CMT

[arXiv]

CMT reduces the training cost of diffusion-based flow map models by up to 90% while reaching SOTA performance

CODA

[arXiv]

Improved object-centric diffusion learning with registers and contrastive alignment

Improved CFG

[arXiv]

An improved mechanism for applying classifier-free guidance in discrete diffusion

SONA

[arXiv]

Learning conditional, unconditional, and matching-aware discriminator with adaptive weighting mechanism (cSAN)

G2D2

[arXiv]

Leveraging discrete diffusion models as priors for inverse problems

TLoRA

[arXiv]

Propose tensor-decomposition-based PEFT method, showing its effectiveness on T-to-I generation tasks

Di4C

[arXiv] [code]

Theoretical analysis of limitation of current discrete diffusion and a method for effectively capturing element-wise dependency

VCT

[arXiv] [code]

Improving Consistency Training with a learned data-noise coupling

Memorization

[arXiv] [code]

Classifier-Free Guidance inside the Attraction Basin May Cause Memorization

Jump Your Steps

[arXiv]

A general method to find an optimal sampling schedule for inference in discrete diffusion

HERO-DM

[arXiv] [demo]

A method efficiently leverages online human feedback to fine-tune Stable Diffusion for various range of tasks

WPSE

[arXiv]

An enhanced multimodal representation using weighted point clouds and its theoretical benefits

PaGoDA

[arXiv]

A 64x64 pre-trained diffusion model is all you need for 1-step high-resolution SOTA generation

CTM

[arXiv] [demo]

Unified framework enables diverse samplers and 1-step generation SOTAs

Applications:
[SoundGen]

SAN

[arXiv] [code] [demo]

Enhancing GAN with metrizable discriminators

Applications:
[Vocoder]

MPGD

[arXiv] [demo]

Fast, Efficient, Training-Free, and Controllable diffusion-based generation method

HQ-VAE

[OpenReview] [arXiv]

Generalizing hierarchical VQ-VAEs with a Bayesian framework

FP-Diffusion

[PMLR] [code]

Improving density estimation of diffusion

GibbsDDRM

[PMLR] [code]

Achieving blind inversion using DDPM

Applications:
[DeReverb] [SpeechEnhance]

SQ-VAE

[PMLR] [arXiv] [code]

Improving codebook utilization and training stability

AR-ELBO

[Elsevier] [arXiv]

Mitigating oversmoothness in VAE

LLM2Fx-Tools

[arXiv] [demo]

Tool Calling For Music Post-Production

MEGAMI

[arXiv] [code] [demo]

Automatic music mixing using a generative model of effect embeddings

Sampling Identification

[arXiv] [code]

Automatic Music Sample Identification with Multi-Track Contrastive Learning

Lyrics Matching

[arXiv] [code]

Leveraging Whisper Embeddings for Audio-based Lyrics Matching

Training Data Attribution

[arXiv]

Large-Scale Training Data Attribution for Music Generative Models via Unlearning

Beyond GenAI Music

[url]

Reductive, exclusionary, normalising: The limits of generative AI music

LLM2FX

[arXiv] [code] [demo] [dataset]

Can Large Language Models Predict Audio Effects Parameters from Natural Language?

Vocal Effects Style Transfer

[arXiv] [code] [demo]

Inference-Time Optimisation for Vocal Effects Style Transfer using DiffVox

Fx-Encoder++

[arXiv] [code]

SOTA Fx representation: Extract instrument-wise audio effects representations from music mixtures

ITO-Master

[arXiv] [code] [demo]

Inference Time Optimization for Music Mastering Style Transfer

GRAFx (ext.)

[JAES] [code] [demo]

Reverse Engineering of Music Mixing Graphs with Differentiable Processors and Iterative Pruning

CLEWS

[arXiv]

Supervised contrastive learning from weakly-labeled audio segments for musical version matching

MFM as Generic Booster

[OpenReview] [arXiv]

Music Foundation Model as Generic Booster for Music Downstream Tasks

DiffVox

[arXiv] [code] [demo] [audio]

DiffVox: A Differentiable Model for Capturing and Analysing Professional Effects Distributions

Variable Bitrate RVQ

[arXiv]

VRVQ: Variable Bitrate Residual Vector Quantization for Audio Compression

Instr. Timbre Transfer

[arXiv] [code] [demo]  

Latent Diffusion Bridges for Unsupervised Musical Audio Timbre Transfer

Mixing Graph Estimation

[arXiv] [code] [demo]

Searching For Music Mixing Graphs: A Pruning Approach

Guitar Amp. Modeling

[arXiv]

Improving Unsupervised Clean-to-Rendered Guitar Tone Transformation Using GANs and Integrated Unaligned Clean Data

Text-to-Music Editing

[arXiv] [code] [demo]

MusicMagus: Zero-Shot Text-to-Music Editing via Diffusion Models

Instr.-Agnostic Trans.

[IEEE] [arXiv]

Timbre-Trap: A Low-Resource Framework for Instrument-Agnostic Music Transcription

Vocal Restoration

[IEEE] [arXiv] [demo]

VRDMG: Vocal Restoration via Diffusion Posterior Sampling with Multiple Guidance

hFT-Transformer

[arXiv] [code]

Automatic Piano Transcription with Hierarchical Frequency-Time Transformer

Automatic Music Tagging

[arXiv]

An Attention-based Approach To Hierarchical Multi-label Music Instrument Classification

Vocal Dereverberation

[arXiv] [demo]

Unsupervised Vocal Dereverberation with Diffusion-based Generative Models

Mixing Style Transfer

[arXiv] [code] [demo]

Music Mixing Style Transfer: A Contrastive Learning Approach to Disentangle Audio Effects

Music Transcription

[arXiv] [code] [demo]

DiffRoll: Diffusion-based Generative Music Transcription with Unsupervised Pretraining Capability

Singing Voice Vocoder

[arXiv] [demo]

Hierarchical Diffusion Models for Singing Voice Neural Vocoder

Distortion Effect Removal

[poster] [arXiv] [demo]

Distortion Audio Effects: Learning How to Recover the Clean Signal

Automatic Music Mixing

[poster] [arXiv] [code] [demo]

Automatic Music Mixing with Deep Learning and Out-of-Domain Data

Sound Separation

[IEEE]

Music Source Separation with Deep Equilibrium Models

Automatic DJ Transition

[arXiv] [code] [demo]

Automatic DJ Transitions with Differentiable Audio Effects and Generative Adversarial Networks

Singing Voice Conversion

[arXiv] [demo]

Robust One-Shot Singing Voice Conversion

Sound Separation

[video] [site]

Glenn Gould and Kanji Ishimaru 2021: A collaboration with AI Sound Separation after 60 years

OmniUE

[arXiv] [demo]

Omni-Interactive Universal Embedder

ST-AudioLM

[arXiv]

Spatio-Temporal Audio Language Modeling for Dynamic Sound Sources

Odoriko

[arXiv] [demo]

Odoriko: A Shape-Aware Multimodal Diffusion Framework for Human Motion

Spectral Alignment (SPA)

[arXiv] [code]

Spectral Prior for Reducing Exposure Bias in Diffusion Models

Step-by-Step V2A

[arXiv] [demo]

Step-by-Step Video-to-Audio Synthesis via Negative Audio Guidance

PAVAS

[CVF] [arXiv] [demo]

PAVAS: a framework for generating physically plausible audio from video, by integrating physics estimation

Echoes Over Time

[CVF] [arXiv] [demo]

Echoes Over Time: Unlocking Length Generalization in Video-to-Audio Generation Models

VIRTUE

[OpenReview] [arXiv] [code] [dataset] [collection]

VIRTUE: Visual-Interactive Text-Image Universal Embedder

Diffiner for SE & SS

[IEEE] [arXiv]

A diffusion-based post-processor for perceptually improving speech enhancement and separation outputs

CCStereo

[ACM] [arXiv] [code]

CCStereo: Audio-Visual Contextual and Contrastive Learning for Binaural Audio Generation

TITAN-Guide

[CVF] [arXiv] [code] [demo]

TITAN-Guide: Taming Inference-Time AligNment for Guided Text-to-Video Diffusion Models

MMAudio

[CVF] [arXiv] [code] [demo]

MMAudio: Taming Multimodal Joint Training for High-Quality Video-to-Audio Synthesis

MMDisCo

[OpenReview] [arXiv] [code]

MMDisCo: Multi-Modal Discriminator-Guided Cooperative Diffusion for Joint Audio and Video Generation

SoundCTM

[OpenReview] [arXiv] [code] [demo]

SoundCTM: Unifying Score-based and Consistency Models for Full-band Text-to-Sound Generation

Mining Your Own Secrets

[OpenReview] [arXiv]

Mining Your Own Secrets: Diffusion Classifier Scores for Continual Personalization of Text-to-Image Diffusion Models

GenWarp

[arXiv] [demo]

GenWarp: Single Image to Novel Views with Semantic-Preserving Generative Warping

SpecMaskGIT

[arXiv] [demo]

SpecMaskGIT: Masked Generative Modeling of Audio Spectrograms for Efficient Audio Synthesis and Beyond

Acoustic Inv. Rendering

[CVF] [arXiv] [dataset] [code] [demo]

Hearing Anything Anywhere

BigVSAN Vocoder

[arXiv] [code] [demo]

BigVSAN: Enhancing GAN-based Neural Vocoders with Slicing Adversarial Network

Zero-/Few-shot SELD

[IEEE] [arXiv]

Zero- and Few-shot Sound Event Localization and Detection

STARSS23

[arXiv] [dataset]

STARSS23: An Audio-Visual Dataset of Spatial Recordings of Real Scenes with Spatiotemporal Annotations of Sound Events

Audio Restoration: ViT-AE

[IEEE] [arXiv] [demo]

Extending Audio Masked Autoencoders Toward Audio Restoration

Diffiner

[ISCA] [arXiv] [code]

Diffiner: A Versatile Diffusion-based Generative Refiner for Speech Enhancement

CLIPSep

[OpenReview] [arXiv] [code] [demo]

CLIPSep: Learning Text-queried Sound Separation with Noisy Unlabeled Videos

Sound Event Localization and Detection

[IEEE] [arXiv]

Multi-ACCDOA: Localizing and Detecting Overlapping Sounds from the Same Class with Auxiliary Duplicating Permutation Invariant Training

DCASE Challenge Task 3

[DCASE Challenge2026]

Semantic Acoustic Imaging for Sound Event Localization and Detection from Spatial Audio and Audiovisual Scenes

CPD Challenge 2025

[CPD Challenge 2025]

Commonsense Persona-grounded Dialogue Challenge 2025

SVG Challenge 2024

[SVG Challenge 2024]

Sounding Video Generation Challenge 2024

CPD Challenge 2023

[CPD Challenge 2023]

Commonsense Persona-grounded Dialogue Challenge

SDX Challenge 2023

[site] [paper (music)] [paper (cinematic)]

Sound Demixing Challenge 2023

MDX Challenge 2021

[site] [frontiers]

Music Demixing Challenge 2021