Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.03531 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917187782967296 |
|---|---|
| author | Wang, Yuwen Qian, Xinyuan Zhang, Tian-Hao Gao, Jiaran Pan, Yuchen Wang, Xin Pan, Zhou Wei, Chen Wang, Yiming |
| author_facet | Wang, Yuwen Qian, Xinyuan Zhang, Tian-Hao Gao, Jiaran Pan, Yuchen Wang, Xin Pan, Zhou Wei, Chen Wang, Yiming |
| contents | Large Audio-Language Models (LALMs) have demonstrated strong performance in audio understanding and generation. Yet, our extensive benchmarking reveals that their behavior is largely generic (e.g., summarizing spoken content) and fails to adequately support personalized question answering (e.g., summarizing what my best friend says). In contrast, human conditions their interpretation and decision-making on each individual's personal context. To bridge this gap, we formalize the task of Personalized LALMs (PALM) for recognizing personal concepts and reasoning within personal context. Moreover, we create the first benchmark (PALM-Bench) to foster the methodological advances in PALM and enable structured evaluation on several tasks across multi-speaker scenarios. Our extensive experiments on representative open-source LALMs, show that existing training-free prompting and supervised fine-tuning strategies, while yield improvements, remains limited in modeling personalized knowledge and transferring them across tasks robustly. Data and code will be released. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_03531 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PALM-Bench: A Comprehensive Benchmark for Personalized Audio-Language Models Wang, Yuwen Qian, Xinyuan Zhang, Tian-Hao Gao, Jiaran Pan, Yuchen Wang, Xin Pan, Zhou Wei, Chen Wang, Yiming Computation and Language Large Audio-Language Models (LALMs) have demonstrated strong performance in audio understanding and generation. Yet, our extensive benchmarking reveals that their behavior is largely generic (e.g., summarizing spoken content) and fails to adequately support personalized question answering (e.g., summarizing what my best friend says). In contrast, human conditions their interpretation and decision-making on each individual's personal context. To bridge this gap, we formalize the task of Personalized LALMs (PALM) for recognizing personal concepts and reasoning within personal context. Moreover, we create the first benchmark (PALM-Bench) to foster the methodological advances in PALM and enable structured evaluation on several tasks across multi-speaker scenarios. Our extensive experiments on representative open-source LALMs, show that existing training-free prompting and supervised fine-tuning strategies, while yield improvements, remains limited in modeling personalized knowledge and transferring them across tasks robustly. Data and code will be released. |
| title | PALM-Bench: A Comprehensive Benchmark for Personalized Audio-Language Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2601.03531 |