Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.23994 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910230958309376 |
|---|---|
| author | Xie, Tianxin Lei, Wentao Jiang, Kai Huang, Guanjie Zhang, Pengfei Zhang, Chunhui Ma, Fengji He, Haoyu Zhang, Han He, Jiangshan Wang, Jinting Fang, Linghan Gao, Lufei Ablet, Orkesh Zhang, Peihua Hu, Ruolin Li, Shengyu Lin, Weilin Feng, Xiaoyang Yang, Xinyue Rong, Yan Wang, Yanyun Shao, Zihang Zhao, Zelin Li, Chenxing Yang, Shan Wang, Wenfu Yu, Meng Yu, Dong Liu, Li |
| author_facet | Xie, Tianxin Lei, Wentao Jiang, Kai Huang, Guanjie Zhang, Pengfei Zhang, Chunhui Ma, Fengji He, Haoyu Zhang, Han He, Jiangshan Wang, Jinting Fang, Linghan Gao, Lufei Ablet, Orkesh Zhang, Peihua Hu, Ruolin Li, Shengyu Lin, Weilin Feng, Xiaoyang Yang, Xinyue Rong, Yan Wang, Yanyun Shao, Zihang Zhao, Zelin Li, Chenxing Yang, Shan Wang, Wenfu Yu, Meng Yu, Dong Liu, Li |
| contents | Text-to-audio-video (T2AV) generation is central to applications such as filmmaking and world modeling. However, current models often fail to produce physically plausible sounds. Previous benchmarks primarily focus on audio-video temporal synchronization, while largely overlooking explicit evaluation of audio-physics grounding, thereby limiting the study of physically plausible audio-visual generation. To address this issue, we present PhyAVBench, the first benchmark that systematically evaluates the audio-physics grounding capabilities of T2AV, image-to-audio-video (I2AV), and video-to-audio (V2A) models. PhyAVBench offers PhyAV-Sound-11K, a new dataset of 25.5 hours of 11,605 audible videos collected from 184 participants to ensure diversity and avoid data leakage. It contains 337 paired-prompt groups with controlled physical variations that drive sound differences, each grounded with an average of 17 videos and spanning 6 audio-physics dimensions and 41 fine-grained test points. Each prompt pair is annotated with the physical factors underlying their acoustic differences. Importantly, PhyAVBench leverages paired text prompts to evaluate this capability. We term this evaluation paradigm the Audio-Physics Sensitivity Test (APST) and introduce a novel metric, the Contrastive Physical Response Score (CPRS), which quantifies the acoustic consistency between generated videos and their real-world counterparts. We conduct a comprehensive evaluation of 17 state-of-the-art models. Our results reveal that even leading commercial models struggle with fundamental audio-physical phenomena, exposing a critical gap beyond audio-visual synchronization and pointing to future research directions. We hope PhyAVBench will serve as a foundation for advancing physically grounded audio-visual generation. Prompts, ground-truth, and generated video samples are available at https://github.com/imxtx/PhyAVBench. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_23994 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PhyAVBench: A Challenging Audio Physics-Sensitivity Benchmark for Physically Grounded Text-to-Audio-Video Generation Xie, Tianxin Lei, Wentao Jiang, Kai Huang, Guanjie Zhang, Pengfei Zhang, Chunhui Ma, Fengji He, Haoyu Zhang, Han He, Jiangshan Wang, Jinting Fang, Linghan Gao, Lufei Ablet, Orkesh Zhang, Peihua Hu, Ruolin Li, Shengyu Lin, Weilin Feng, Xiaoyang Yang, Xinyue Rong, Yan Wang, Yanyun Shao, Zihang Zhao, Zelin Li, Chenxing Yang, Shan Wang, Wenfu Yu, Meng Yu, Dong Liu, Li Sound Artificial Intelligence Text-to-audio-video (T2AV) generation is central to applications such as filmmaking and world modeling. However, current models often fail to produce physically plausible sounds. Previous benchmarks primarily focus on audio-video temporal synchronization, while largely overlooking explicit evaluation of audio-physics grounding, thereby limiting the study of physically plausible audio-visual generation. To address this issue, we present PhyAVBench, the first benchmark that systematically evaluates the audio-physics grounding capabilities of T2AV, image-to-audio-video (I2AV), and video-to-audio (V2A) models. PhyAVBench offers PhyAV-Sound-11K, a new dataset of 25.5 hours of 11,605 audible videos collected from 184 participants to ensure diversity and avoid data leakage. It contains 337 paired-prompt groups with controlled physical variations that drive sound differences, each grounded with an average of 17 videos and spanning 6 audio-physics dimensions and 41 fine-grained test points. Each prompt pair is annotated with the physical factors underlying their acoustic differences. Importantly, PhyAVBench leverages paired text prompts to evaluate this capability. We term this evaluation paradigm the Audio-Physics Sensitivity Test (APST) and introduce a novel metric, the Contrastive Physical Response Score (CPRS), which quantifies the acoustic consistency between generated videos and their real-world counterparts. We conduct a comprehensive evaluation of 17 state-of-the-art models. Our results reveal that even leading commercial models struggle with fundamental audio-physical phenomena, exposing a critical gap beyond audio-visual synchronization and pointing to future research directions. We hope PhyAVBench will serve as a foundation for advancing physically grounded audio-visual generation. Prompts, ground-truth, and generated video samples are available at https://github.com/imxtx/PhyAVBench. |
| title | PhyAVBench: A Challenging Audio Physics-Sensitivity Benchmark for Physically Grounded Text-to-Audio-Video Generation |
| topic | Sound Artificial Intelligence |
| url | https://arxiv.org/abs/2512.23994 |