analiza aiqbench
ToolDF introduces a novel framework using audio large language models to orchestrate specialized tools for detecting deepfakes with mixed authenticity, addressing the complex reality where genuine and manipulated cues coexist in a single clip. By adaptively routing audio components to domain-specific experts, this approach achieves significant performance gains over monolithic baselines while providing interpretable evidence localized to specific temporal regions and sources. This work advances the field of AI safety by moving beyond simple binary classification toward robust, explainable detection systems capable of handling hybrid acoustic mixtures.
ToolDF: Tool-Integrated Reasoning for Mixed-Authenticity Audio Deepfake Detection
arXiv:2609.03620v1 Announce Type: cross Abstract: Audio deepfake detection is commonly formulated as clip-level binary classification of single-domain audio. However, real-world manipulated audio can exhibit mixed authenticity, where genuine and manipulated cues coexist across temporal transitions, overlapping sources, or both. This setting requires not only detecting manipulated audio but also localizing the components that provide evidence for the decision. We propose ToolDF, a tool-integrated reasoning framework for mixed-authenticity audio deepfake detection. ToolDF employs an audio large language model as an orchestrator trained with supervised tool-use trajectories. It adaptively analyzes the audio scene, selectively performs source separation, routes components to domain-specific experts, and aggregates their evidence into an interpretable verdict. We further introduce a mixed-authenticity ADD benchmark covering temporal transitions, acoustic overlaps, and hybrid mixtures. Experimental results show that ToolDF achieves the best overall performance on composite-type detection, achieving macro-F1 gains of 3.72 and 14.39 points over the strongest monolithic baseline and a fixed pipeline, respectively, while providing interpretable evidence localized to temporal regions and acoustic sources. Our source code and dataset are publicly available online.
Abstract:Audio deepfake detection is commonly formulated as clip-level binary classification of single-domain audio. However, real-world manipulated audio can exhibit mixed authenticity, where genuine and manipulated cues coexist across temporal transitions, overlapping sources, or both. This setting requires not only detecting manipulated audio but also localizing the components that provide evidence for the decision. We propose ToolDF, a tool-integrated reasoning framework for mixed-authenticity audio deepfake detection. ToolDF employs an audio large language model as an orchestrator trained with supervised tool-use trajectories. It adaptively analyzes the audio scene, selectively performs source separation, routes components to domain-specific experts, and aggregates their evidence into an interpretable verdict. We further introduce a mixed-authenticity ADD benchmark covering temporal transitions, acoustic overlaps, and hybrid mixtures. Experimental results show that ToolDF achieves the best overall performance on composite-type detection, achieving macro-F1 gains of 3.72 and 14.39 points over the strongest monolithic baseline and a fixed pipeline, respectively, while providing interpretable evidence localized to temporal regions and acoustic sources. Our source code and dataset are publicly available online.
Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI); Sound (cs.SD)
arXiv:2609.03620 [eess.AS]
(or arXiv:2609.03620v1 [eess.AS] for this version)
https://doi.org/10.48550/arXiv.2609.03620 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 270 cuvinte
Subiecte conexe
[AINews] The Future of Latent Space
Latent Space · 25 sept. 2026
A Candid Abacus AI Review: The All-in-One AI Platform for Professionals & Enterprises
KDnuggets · 10 sept. 2026
[AINews] High Taste Testers Ep. 1
Latent Space · 17 oct. 2026
2026.9: There's room on this bus
Home Assistant · 2 sept. 2026