AHELM standardizes evaluation of audio-language models across 10 aspects and shows simple ASR+LLM systems are competitive with multimodal models.
MuTox: Universal MUltilingual Audio-based TOXicity Dataset and Zero-shot Detector
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Research in toxicity detection in natural language processing for the speech modality (audio-based) is quite limited, particularly for languages other than English. To address these limitations and lay the groundwork for truly multilingual audio-based toxicity detection, we introduce MuTox, the first highly multilingual audio-based dataset with toxicity labels. The dataset comprises 20,000 audio utterances for English and Spanish, and 4,000 for the other 19 languages. To demonstrate the quality of this dataset, we trained the MuTox audio-based toxicity classifier, which enables zero-shot toxicity detection across a wide range of languages. This classifier outperforms existing text-based trainable classifiers by more than 1% AUC, while expanding the language coverage more than tenfold. When compared to a wordlist-based classifier that covers a similar number of languages, MuTox improves precision and recall by approximately 2.5 times. This significant improvement underscores the potential of MuTox in advancing the field of audio-based toxicity detection.
citation-role summary
citation-polarity summary
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
AHELM: A Holistic Evaluation of Audio-Language Models
AHELM standardizes evaluation of audio-language models across 10 aspects and shows simple ASR+LLM systems are competitive with multimodal models.