The Interspeech 2026 Audio Reasoning Challenge: Evaluating Reasoning Process Quality for Audio Reasoning Models and Agents
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917275602255872 |
|---|---|
| author | Ma, Ziyang Xu, Ruiyang Ma, Yinghao Yang, Chao-Han Huck Li, Bohan Kim, Jaeyeon Xu, Jin Li, Jinyu Busso, Carlos Yu, Kai Chng, Eng Siong Chen, Xie |
| author_facet | Ma, Ziyang Xu, Ruiyang Ma, Yinghao Yang, Chao-Han Huck Li, Bohan Kim, Jaeyeon Xu, Jin Li, Jinyu Busso, Carlos Yu, Kai Chng, Eng Siong Chen, Xie |
| contents | Recent Large Audio Language Models (LALMs) excel in understanding but often lack transparent reasoning. To address this "black-box" limitation, we organized the Audio Reasoning Challenge at Interspeech 2026, the first shared task dedicated to evaluating Chain-of-Thought (CoT) quality in the audio domain. The challenge introduced MMAR-Rubrics, a novel instance-level protocol assessing the factuality and logic of reasoning chains. Featured Single Model and Agent tracks, the competition attracting 156 teams from 18 countries and regions. Results show agent systems currently lead in reasoning quality, utilizing iterative tool orchestration and cross-modal analysis. Besides, single models are rapidly advancing via reinforcement learning and sophisticated data pipeline. We details the challenge design, methodology, and a comprehensive analysis of state-of-the-art systems, providing new insights for explainable audio intelligence. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_14224 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | The Interspeech 2026 Audio Reasoning Challenge: Evaluating Reasoning Process Quality for Audio Reasoning Models and Agents Ma, Ziyang Xu, Ruiyang Ma, Yinghao Yang, Chao-Han Huck Li, Bohan Kim, Jaeyeon Xu, Jin Li, Jinyu Busso, Carlos Yu, Kai Chng, Eng Siong Chen, Xie Sound Computation and Language Multimedia Recent Large Audio Language Models (LALMs) excel in understanding but often lack transparent reasoning. To address this "black-box" limitation, we organized the Audio Reasoning Challenge at Interspeech 2026, the first shared task dedicated to evaluating Chain-of-Thought (CoT) quality in the audio domain. The challenge introduced MMAR-Rubrics, a novel instance-level protocol assessing the factuality and logic of reasoning chains. Featured Single Model and Agent tracks, the competition attracting 156 teams from 18 countries and regions. Results show agent systems currently lead in reasoning quality, utilizing iterative tool orchestration and cross-modal analysis. Besides, single models are rapidly advancing via reinforcement learning and sophisticated data pipeline. We details the challenge design, methodology, and a comprehensive analysis of state-of-the-art systems, providing new insights for explainable audio intelligence. |
| title | The Interspeech 2026 Audio Reasoning Challenge: Evaluating Reasoning Process Quality for Audio Reasoning Models and Agents |
| topic | Sound Computation and Language Multimedia |
| url | https://arxiv.org/abs/2602.14224 |