Ghost in the Encoder: Decodable Artist Identity Representations in Lyrics-to-Song Generation
cs.SD, cs.LG
Submitted: 2026-09-30
Updated: 2026-09-30
Code: https://github.com/ace-step/ACE-Step-1.5
Project page: https://ismir-2026-gine.github.io/ghost-inthe-encoder
Sources
- Bob's Confetti: Phonetic Memorization Attacks in Music and Video Generation
- YuE: Scaling Open Foundation Models for Long-Form Music Generation
- ACE-Step 1.5: Pushing the Boundaries of Open-Source Music Generation
- SongGen: A Single Stage Auto-regressive Transformer for Text-to-Song Generation
- DiffRhythm: Blazingly Fast and Embarrassingly Simple End-to-End Full-Length Song Generation with Latent Diffusion
- ACE-Step: A Step Towards Music Generation Foundation Model
- Jukebox: A Generative Model for Music
- Long-form music generation with latent diffusion
- Qwen3 Technical Report
Related papers
- Few-Shot Open-Set Audio Classification via Transductive Prototype Refinement and Class Logit Enhancement
- Spectral Masking and Interpolation Attack (SMIA): A Black-box Adversarial Attack against Voice Authentication and Anti-Spoofing Systems
- Remix the Timbre: Diffusion-Based Style Transfer Across Polyphonic Stems
- AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and Thinking
- SoundWeaver: Compositional Warm-Starting for Text-to-Audio Diffusion Serving
- WASIL: In-the-Wild Arabic Spoken Interactions with LLMs