安定報道日時 2026-10-02 12:00
合成音声研究対象の監査に新しい方法
新しい合成音声研究対象のサブストレートが提案され、生成プロヴァンス、ソース指定、品質信号を含み、モデルの挙動を属性付ける能力を向上させています。
01
根拠
- AarXiv cs.AI一次情報2026-10-02 12:00
> Abstract: Attributing model behavior to synthetic training data requires knowing what produced each training item before estimating what that item caused.
- AarXiv cs.AI一次情報2026-10-02 12:00
Abstract: Attributing model behavior to synthetic training data requires knowing what produced each training item before estimating what that item caused. A waveform-label pair does not preserve this knowledge. We propose a generation-provenance substrate in which a synthetic research object binds source specification, generated content, waveform, target, fact requirements, quality signals, revie…
- AarXiv cs.AI一次情報2026-10-02 12:00
Comments: Accepted to the Third NeurIPS Workshop on Attributing Model Behavior at Scale: Data Attribution and Provenance. 4 pages, 0 figures, 1 table.