Benchmarking LLMs on the Massive Sound Embedding Benchmark (MSEB)
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918485935783936 |
|---|---|
| author | Allauzen, Cyril Bagby, Tom Heigold, Georg Variani, Ehsan Wu, Ke |
| author_facet | Allauzen, Cyril Bagby, Tom Heigold, Georg Variani, Ehsan Wu, Ke |
| contents | The Massive Sound Embedding Benchmark (MSEB) has emerged as a standard for evaluating the functional breadth of audio models. While initial baselines focused on specialized encoders, the shift toward "audio-native" Large Language Models (LLMs) suggests a new paradigm where a single multimodal backbone may replace complex, task-specific pipelines. This paper provides a rigorous empirical evaluation of leading LLMs - including members from the Gemini and GPT families - across the eight core MSEB capabilities to assess their efficacy and audio-text parity. Our results indicate that while a significant modality gap persists regarding performance and robustness, the empirical evidence for an "optimal" modeling approach remains inconclusive. Ultimately, the choice between audionative and cascaded architectures depends heavily on specific use-case requirements and the underlying assumptions regarding latency, cost, and reasoning depth. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_04556 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Benchmarking LLMs on the Massive Sound Embedding Benchmark (MSEB) Allauzen, Cyril Bagby, Tom Heigold, Georg Variani, Ehsan Wu, Ke Sound Machine Learning The Massive Sound Embedding Benchmark (MSEB) has emerged as a standard for evaluating the functional breadth of audio models. While initial baselines focused on specialized encoders, the shift toward "audio-native" Large Language Models (LLMs) suggests a new paradigm where a single multimodal backbone may replace complex, task-specific pipelines. This paper provides a rigorous empirical evaluation of leading LLMs - including members from the Gemini and GPT families - across the eight core MSEB capabilities to assess their efficacy and audio-text parity. Our results indicate that while a significant modality gap persists regarding performance and robustness, the empirical evidence for an "optimal" modeling approach remains inconclusive. Ultimately, the choice between audionative and cascaded architectures depends heavily on specific use-case requirements and the underlying assumptions regarding latency, cost, and reasoning depth. |
| title | Benchmarking LLMs on the Massive Sound Embedding Benchmark (MSEB) |
| topic | Sound Machine Learning |
| url | https://arxiv.org/abs/2605.04556 |