Image de couverture de livre

Training data attribution in diffusion models via mirrored unlearning and noise-consistent skew

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Serrà, Joan, Goswami, Dipam, Morreale, Fabio, Liao, Wei-Hsiang, Mitsufuji, Yuki
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Emergent, not Immanent: A Baradian Reading of Explainable AI
par: Morreale, Fabio, et autres
Publié: (2026)

Attribution-by-design: Ensuring Inference-Time Provenance in Generative Music Systems
par: Morreale, Fabio, et autres
Publié: (2025)

Automatic Music Sample Identification with Multi-Track Contrastive Learning
par: Riou, Alain, et autres
Publié: (2025)

A Comprehensive Real-World Assessment of Audio Watermarking Algorithms: Will They Survive Neural Codecs?
par: Özer, Yigitcan, et autres
Publié: (2025)

Leveraging Whisper Embeddings for Audio-based Lyrics Matching
par: Mancini, Eleonora, et autres
Publié: (2025)

Supervised contrastive learning from weakly-labeled audio segments for musical version matching
par: Serrà, Joan, et autres
Publié: (2025)

Provable unlearning in topic modeling and downstream tasks
par: Wei, Stanley, et autres
Publié: (2024)

Applying sparse autoencoders to unlearn knowledge in language models
par: Farrell, Eoin, et autres
Publié: (2024)

Towards Assessing Data Replication in Music Generation with Music Similarity Metrics on Raw Audio
par: Batlle-Roca, Roser, et autres
Publié: (2024)

MR-MT3: Memory Retaining Multi-Track Music Transcription to Mitigate Instrument Leakage
par: Tan, Hao Hao, et autres
Publié: (2024)

Machine unlearning through fine-grained model parameters perturbation
par: Zuo, Zhiwei, et autres
Publié: (2024)

GUDA: Counterfactual Group-wise Training Data Attribution for Diffusion Models via Unlearning
par: Murata, Naoki, et autres
Publié: (2026)

Woosh: A Sound Effects Foundation Model
par: Hadjeres, Gaëtan, et autres
Publié: (2026)

PaGoDA: Progressive Growing of a One-Step Generator from a Low-Resolution Diffusion Teacher
par: Kim, Dongjun, et autres
Publié: (2024)

CMT: Mid-Training for Efficient Learning of Consistency, Mean Flow, and Flow Map Models
par: Hu, Zheyuan, et autres
Publié: (2025)

On the limitation of evaluating machine unlearning using only a single training seed
par: Lanyon, Jamie, et autres
Publié: (2025)

Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion
par: Kim, Dongjun, et autres
Publié: (2023)

Instruct-MusicGen: Unlocking Text-to-Music Editing for Music Language Models via Instruction Tuning
par: Zhang, Yixiao, et autres
Publié: (2024)

Concept-TRAK: Understanding how diffusion models learn concepts through concept-level attribution
par: Park, Yonghyun, et autres
Publié: (2025)

Distill, Forget, Repeat: A Framework for Continual Unlearning in Text-to-Image Diffusion Models
par: George, Naveen, et autres
Publié: (2025)

Edge-preserving noise for diffusion models
par: Vandersanden, Jente, et autres
Publié: (2024)

Opening Musical Creativity? Embedded Ideologies in Generative-AI Music Systems
par: Pram, Liam, et autres
Publié: (2025)

Understanding and Accelerating the Training of Masked Diffusion Language Models
par: Hong, Chunsan, et autres
Publié: (2026)

Step-resolved data attribution for looped transformers
par: Kaissis, Georgios, et autres
Publié: (2026)

The Principles of Diffusion Models
par: Lai, Chieh-Hsin, et autres
Publié: (2025)

ROCM: RLHF on consistency models
par: Shekhar, Shivanshu, et autres
Publié: (2025)

Federated Learning for distribution skewed data using sample weights
par: Nguyen, Hung, et autres
Publié: (2024)

Improving Classifier-Free Guidance in Masked Diffusion: Low-Dim Theoretical Insights with High-Dim Impact
par: Rojas, Kevin, et autres
Publié: (2025)

HQ-VAE: Hierarchical Discrete Representation Learning with Variational Bayes
par: Takida, Yuhta, et autres
Publié: (2023)

It`s All About Speed: AI`s Impact on Workflow in Music Production
par: McClellan, Finn, et autres
Publié: (2026)

Routing without Forgetting
par: Masano, Alessio, et autres
Publié: (2026)

Denoising Multi-Beta VAE: Representation Learning for Disentanglement and Generation
par: Uppal, Anshuk, et autres
Publié: (2025)

DisMix: Disentangling Mixtures of Musical Instruments for Source-level Pitch and Timbre Manipulation
par: Luo, Yin-Jyun, et autres
Publié: (2024)

HERO: Human-Feedback Efficient Reinforcement Learning for Online Diffusion Model Finetuning
par: Hiranaka, Ayano, et autres
Publié: (2024)

Wind speed super-resolution and validation: from ERA5 to CERRA via diffusion models
par: Merizzi, Fabio, et autres
Publié: (2024)

DUET: Optimizing Training Data Mixtures via Feedback from Unseen Evaluation Tasks
par: Chen, Zhiliang, et autres
Publié: (2025)

Fixing confirmation bias in feature attribution methods via semantic match
par: Cinà, Giovanni, et autres
Publié: (2023)

Generalization through variance: how noise shapes inductive biases in diffusion models
par: Vastola, John J.
Publié: (2025)

MusicMagus: Zero-Shot Text-to-Music Editing via Diffusion Models
par: Zhang, Yixiao, et autres
Publié: (2024)

Deliberative Alignment is Deep, but Uncertainty Remains: Inference time safety improvement in reasoning via attribution of unsafe behavior to base model
par: Pathmanathan, Pankayaraj, et autres
Publié: (2026)