Existing audio forgery detection and localization (AFDL) methods often overfit dataset-specific low-level artifacts, limiting their generalization to subtle, localized, and unseen manipulations. Re...
<p><strong>Tác giả:</strong> Yuxiong Xu, Kaiqing Lin, Bin Li, Haodong Li, Sheng Li</p>
<p>Existing audio forgery detection and localization (AFDL) methods often overfit dataset-specific low-level artifacts, limiting their generalization to subtle, localized, and unseen manipulations. Recent audio large language model (ALLM)-based approaches cast AFDL as question answering but still model forensic evidence implicitly, without linking manipulation cues to predictions. To bridge this gap, we propose ThinkOmni, a reasoning-driven omni-modal large language model that jointly performs e...</p>
<p>📄 DOI: <a href="https://doi.org/https://doi.org/10.1145/3767308.3835017" target="_blank" rel="noopener">https://doi.org/10.1145/3767308.3835017</a></p>
<p>📖 <a href="https://doi.org/https://doi.org/10.1145/3767308.3835017" target="_blank" rel="noopener">Xem bài báo đầy đủ</a></p>
Bài viết được biên soạn lại bởi LIFVietnam. Nguồn tham khảo: OpenAlex
