← Back to all news
arXiv Machine Learning By Gang Shi (Independent Researcher)

Clean Accuracy Does Not Guarantee Provenance Robustness: A Prospective Codec-Stress Evaluation of Audio Attribution

Read the original on arXiv Machine Learning →

The study evaluates audio provenance attribution systems, showing that high clean‑benchmark accuracy does not translate to robustness after codec compression. Using a prospectively registered protocol, the authors measured closed‑set attribution performance on two corpora after single‑stage codec transport, finding significant degradation—up to 70.3 Macro‑F1 points for WavLM‑Base+ and 61.0 for W2V2‑BERT 2.0—depending on codec settings and representation. The results demonstrate that clean accuracy alone cannot guarantee deployment robustness across different codecs and representations.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Sep 3

Half-Truth Audio Detection and Localisation: A Lightweight Cross-Attentive Architecture and a Cross-Corpus Diagnostic Study

The paper introduces CAFNet, a lightweight cross‑attentive neural network that fuses MFCC, LFCC, and Chroma‑STFT features to detect and localise partially manipulated (half‑truth) speech. CAFNet achieves high ternary accuracy (97.55%) and low boundary mean absolute error (0.037 s) on the MLADDC benchmark, while demonstrating that cross‑corpus transfer depends on both capability and corpus characteristics. Ablation studies show that cross‑attention fusion is the most critical component, and removing a deeply supervised auxiliary head improves in‑domain performance and reduces variance.

By S. Sutharya, Remya K. Sasi
benchmarks
More like this →
arXiv Computation and Language
Sep 15

Auditing Generative Audio Calls for Known-Task Audio-LLM Evaluatio

The paper investigates how to evaluate generative audio large language models (Audio‑LLMs) on known closed‑set tasks by separating the decision to call a generative model from the use of acoustic evidence. It introduces a controlled call‑decision framework where a policy can choose between a transcript label, encoder evidence from CLAP, AST, or WavLM, or a generative call to Qwen2‑Audio, Qwen2.5‑Omni, or MOSS‑Audio, and measures the impact of generative calls on accuracy. Results on the VocalSound dataset show that while transcript‑only accuracy is low (0.296), encoder‑based controls achieve high accuracy (≈0.85) without any generative calls, and adding generative calls yields only a marginal improvement (0.925 vs. 0.921).

By Mengzhe Geng
llmsmultimodalbenchmarks
More like this →