Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.12186 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913995300012032 |
|---|---|
| author | Chun, Chaewan Terrisse, Lysandre Zhang, Delvin Ce Lee, Dongwon |
| author_facet | Chun, Chaewan Terrisse, Lysandre Zhang, Delvin Ce Lee, Dongwon |
| contents | Despite the growing popularity of audio platforms, fact-checking spoken content remains significantly underdeveloped. Misinformation in speech often unfolds across multi-turn dialogues, shaped by speaker interactions, disfluencies, overlapping speech, and emotional tone-factors that complicate both claim detection and verification. Existing datasets fall short by focusing on isolated sentences or text transcripts, without modeling the conversational and acoustic complexity of spoken misinformation. We introduce MAD (Multi-turn Audio Dialogues), the first fact-checking dataset aligned with multi-turn spoken dialogues and corresponding audio. MAD captures how misinformation is introduced, contested, and reinforced through natural conversation. Each dialogue includes annotations for speaker turns, dialogue scenarios, information spread styles, sentence-level check-worthiness, and both sentence- and dialogue-level veracity. The dataset supports two core tasks: check-worthy claim detection and claim verification. Benchmarking shows that even strong pretrained models reach only 72-74% accuracy at the sentence level and 71-72% at the dialogue level in claim verification, underscoring MAD's difficulty. MAD offers a high-quality benchmark for advancing multimodal and conversational fact-checking, while also surfacing open challenges related to reasoning over speech and dialogue dynamics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_12186 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MAD: A Benchmark for Multi-Turn Audio Dialogue Fact-Checking Chun, Chaewan Terrisse, Lysandre Zhang, Delvin Ce Lee, Dongwon Social and Information Networks Despite the growing popularity of audio platforms, fact-checking spoken content remains significantly underdeveloped. Misinformation in speech often unfolds across multi-turn dialogues, shaped by speaker interactions, disfluencies, overlapping speech, and emotional tone-factors that complicate both claim detection and verification. Existing datasets fall short by focusing on isolated sentences or text transcripts, without modeling the conversational and acoustic complexity of spoken misinformation. We introduce MAD (Multi-turn Audio Dialogues), the first fact-checking dataset aligned with multi-turn spoken dialogues and corresponding audio. MAD captures how misinformation is introduced, contested, and reinforced through natural conversation. Each dialogue includes annotations for speaker turns, dialogue scenarios, information spread styles, sentence-level check-worthiness, and both sentence- and dialogue-level veracity. The dataset supports two core tasks: check-worthy claim detection and claim verification. Benchmarking shows that even strong pretrained models reach only 72-74% accuracy at the sentence level and 71-72% at the dialogue level in claim verification, underscoring MAD's difficulty. MAD offers a high-quality benchmark for advancing multimodal and conversational fact-checking, while also surfacing open challenges related to reasoning over speech and dialogue dynamics. |
| title | MAD: A Benchmark for Multi-Turn Audio Dialogue Fact-Checking |
| topic | Social and Information Networks |
| url | https://arxiv.org/abs/2508.12186 |