MDAR: A Multi-scene Dynamic Audio Reasoning Benchmark

• URL: http://arxiv.org/abs/2509.22461v1
• Date: Fri, 26 Sep 2025 15:12:46 GMT
• Title: MDAR: A Multi-scene Dynamic Audio Reasoning Benchmark
• Authors: Hui Li, Changhao Jiang, Hongyu Wang, Ming Zhang, Jiajun Sun, Zhixiong Yang, Yifei Cao, Shihan Dou, Xiaoran Fan, Baoyu Fan, Tao Ji, Tao Gui, Qi Zhang, Xuanjing Huang,
• Abstract summary: We introduce MDAR, a benchmark for evaluating models on complex, multi-scene, and dynamically evolving audio reasoning tasks.<n>MDAR comprises 3,000 carefully curated question-answer pairs linked to diverse audio clips.<n>We benchmark 26 state-of-the-art audio language models on MDAR and observe that they exhibit limitations in complex reasoning tasks.
• Score: 64.89810922949984
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The ability to reason from audio, including speech, paralinguistic cues, environmental sounds, and music, is essential for AI agents to interact effectively in real-world scenarios. Existing benchmarks mainly focus on static or single-scene settings and do not fully capture scenarios where multiple speakers, unfolding events, and heterogeneous audio sources interact. To address these challenges, we introduce MDAR, a benchmark for evaluating models on complex, multi-scene, and dynamically evolving audio reasoning tasks. MDAR comprises 3,000 carefully curated question-answer pairs linked to diverse audio clips, covering five categories of complex reasoning and spanning three question types. We benchmark 26 state-of-the-art audio language models on MDAR and observe that they exhibit limitations in complex reasoning tasks. On single-choice questions, Qwen2.5-Omni (open-source) achieves 76.67% accuracy, whereas GPT-4o Audio (closed-source) reaches 68.47%; however, GPT-4o Audio substantially outperforms Qwen2.5-Omni on the more challenging multiple-choice and open-ended tasks. Across all three question types, no model achieves 80% performance. These findings underscore the unique challenges posed by MDAR and its value as a benchmark for advancing audio reasoning research.Code and benchmark can be found at https://github.com/luckyerr/MDAR.

Related papers

• JointAVBench: A Benchmark for Joint Audio-Visual Reasoning Evaluation [16.067014259345743]
  We evaluate leading vision-only, audio-only, and Omni-LLMs on our dataset.<n>Even the best-performing Omni-LLM achieves an average accuracy of only 62.6%, outperforming uni-modal baselines.
  arXiv  Detail & Related papers  (2025-12-14T17:23:21Z)
• VoiceAssistant-Eval: Benchmarking AI Assistants across Listening, Speaking, and Viewing [45.15289852736435]
  VoiceAssistant-Eval comprises 10,497 curated examples spanning 13 task categories.<n>To demonstrate its utility, we evaluate 21 open-source models and GPT-4o-Audio.<n>Results reveal three key findings: proprietary models do not universally outperform open-source models.
  arXiv  Detail & Related papers  (2025-09-26T17:59:59Z)
• AHELM: A Holistic Evaluation of Audio-Language Models [78.20477815156484]
  multimodal audio-language models (ALMs) take interleaved audio and text as input and output text.<n>AHELM is a benchmark that aggregates various datasets -- including 2 new synthetic audio-text datasets called PARADE and CoRe-Bench.<n>We also standardize the prompts, inference parameters, and evaluation metrics to ensure equitable comparisons across models.
  arXiv  Detail & Related papers  (2025-08-29T07:40:39Z)
• MAGNET: A Multi-agent Framework for Finding Audio-Visual Needles by Reasoning over Multi-Video Haystacks [67.31276358668424]
  We introduce a novel task named AV-HaystacksQA, where the goal is to identify salient segments across different videos in response to a query and link them together to generate the most informative answer.<n> AVHaystacks is an audio-visual benchmark comprising 3100 annotated QA pairs designed to assess the capabilities of LMMs in multi-video retrieval and temporal grounding task.<n>We propose a model-agnostic, multi-agent framework to address this challenge, achieving up to 89% and 65% relative improvements over baseline methods on BLEU@4 and GPT evaluation scores in QA task on our proposed AVHaystack
  arXiv  Detail & Related papers  (2025-06-08T06:34:29Z)
• Multi-Domain Audio Question Answering Toward Acoustic Content Reasoning in The DCASE 2025 Challenge [102.84031769492708]
  This task defines three QA subsets to test audio-language models on interactive question-answering over diverse acoustic scenes.<n>Preliminary results on the development set are compared, showing strong variation across models and subsets.<n>This challenge aims to advance the audio understanding and reasoning capabilities of audio-language models toward human-level acuity.
  arXiv  Detail & Related papers  (2025-05-12T09:04:16Z)
• Mellow: a small audio language model for reasoning [31.309253699062307]
  Mellow is a small Audio-Language Model specifically designed for reasoning.<n>ReasonAQA is a dataset designed to enhance audio-grounded reasoning in models.<n>Our training dataset, findings, and baseline pave the way for developing small ALMs capable of reasoning.
  arXiv  Detail & Related papers  (2025-03-11T15:29:00Z)
• AV-Odyssey Bench: Can Your Multimodal LLMs Really Understand Audio-Visual Information? [65.49972312524724]
  multimodal large language models (MLLMs) have expanded their capabilities to include vision and audio modalities.<n>Our proposed DeafTest reveals that MLLMs often struggle with simple tasks humans find trivial.<n>We introduce AV-Odyssey Bench, a comprehensive audio-visual benchmark designed to assess whether those MLLMs can truly understand the audio-visual information.
  arXiv  Detail & Related papers  (2024-12-03T17:41:23Z)
• MMAU: A Massive Multi-Task Audio Understanding and Reasoning Benchmark [44.672035866509624]
  MMAU comprises 10k carefully curated audio clips paired with human-annotated natural language questions and answers. It includes information extraction and reasoning questions, requiring models to demonstrate 27 distinct skills across unique and challenging tasks. We assess 18 open-source and proprietary (Large) Audio-Language Models, demonstrating the significant challenges posed by MMAU.
  arXiv  Detail & Related papers  (2024-10-24T21:20:10Z)
• AIR-Bench: Benchmarking Large Audio-Language Models via Generative Comprehension [95.8442896569132]
  We introduce AIR-Bench, the first benchmark to evaluate the ability of Large Audio-Language Models (LALMs) to understand various types of audio signals and interact with humans in the textual format. Results demonstrate a high level of consistency between GPT-4-based evaluation and human evaluation.
  arXiv  Detail & Related papers  (2024-02-12T15:41:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.