SMITIN: Self-Monitored Inference-Time INtervention for Generative Music Transformers
We introduce Self-Monitored Inference-Time INtervention (SMITIN), an approach for controlling an autoregressive generative music transformer using classifier probes. These simple logistic regression probes are trained on the output of each attention head in the transformer using a small dataset of a...
Saved in:
| Main Authors: | Junghyun Koo, Gordon Wichern, Francois G. Germain, Sameer Khurana, Jonathan Le Roux |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Open Journal of Signal Processing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10856829/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Similar Items
-
Music information analysis and retrieval techniques
by: Bożena KOSTEK, et al.
Published: (2008-01-01) -
A Key-Finding Algorithm Based on Music Signature
by: Dariusz KANIA, et al.
Published: (2019-07-01) -
Auditory Display Applied to Research in Music and Acoustics
by: Bożena KOSTEK
Published: (2014-06-01) -
CCMusic: An Open and Diverse Database for Chinese Music Information Retrieval Research
by: Monan Zhou, et al.
Published: (2025-03-01) -
Automatic Music Summarization. A “Thumbnail” Approach
by: Jakub GŁACZYNSKI, et al.
Published: (2013-10-01)