Eureka-Audio: Triggering Audio Intelligence in Compact Language Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhang, Dan, Lei, Yishu, Hu, Jing, He, Shuwei, Deng, Songhe, Luo, Xianlong, Zhu, Danxiang, Feng, Shikun, Liu, Rui, He, Jingzhou, Sun, Yu, Wu, Hua, Wang, Haifeng
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908835278487552
author Zhang, Dan
Lei, Yishu
Hu, Jing
He, Shuwei
Deng, Songhe
Luo, Xianlong
Zhu, Danxiang
Feng, Shikun
Liu, Rui
He, Jingzhou
Sun, Yu
Wu, Hua
Wang, Haifeng
author_facet Zhang, Dan
Lei, Yishu
Hu, Jing
He, Shuwei
Deng, Songhe
Luo, Xianlong
Zhu, Danxiang
Feng, Shikun
Liu, Rui
He, Jingzhou
Sun, Yu
Wu, Hua
Wang, Haifeng
contents We present Eureka-Audio, a compact yet high-performance audio language model that achieves competitive performance against models that are 4 to 18 times larger across a broad range of audio understanding benchmarks. Despite containing only 1.7B parameters, Eureka-Audio demonstrates strong performance on automatic speech recognition (ASR), audio understanding, and dense audio captioning, matching or surpassing multiple 7B to 30B audio and omni-modal baselines. The model adopts a unified end-to-end architecture composed of a lightweight language backbone, a Whisper-based audio encoder, and a sparsely activated Mixture-of-Experts (MoE) adapter that explicitly accounts for audio heterogeneity and alleviates cross-modal optimization conflicts under limited capacity. To further enhance paralinguistic reasoning, we introduce DataFlux, a closed loop audio instruction data synthesis and verification pipeline that constructs high quality, logically consistent supervision from raw audio. Extensive evaluations across ASR, knowledge reasoning, safety, instruction following, and paralinguistic benchmarks, demonstrate that Eureka-Audio achieves an efficient balance between computational cost and performance. These results establish Eureka Audio as a strong and practical baseline for lightweight audio understanding models.
format Preprint
id arxiv_https___arxiv_org_abs_2602_13954
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Eureka-Audio: Triggering Audio Intelligence in Compact Language Models
Zhang, Dan
Lei, Yishu
Hu, Jing
He, Shuwei
Deng, Songhe
Luo, Xianlong
Zhu, Danxiang
Feng, Shikun
Liu, Rui
He, Jingzhou
Sun, Yu
Wu, Hua
Wang, Haifeng
Sound
Artificial Intelligence
We present Eureka-Audio, a compact yet high-performance audio language model that achieves competitive performance against models that are 4 to 18 times larger across a broad range of audio understanding benchmarks. Despite containing only 1.7B parameters, Eureka-Audio demonstrates strong performance on automatic speech recognition (ASR), audio understanding, and dense audio captioning, matching or surpassing multiple 7B to 30B audio and omni-modal baselines. The model adopts a unified end-to-end architecture composed of a lightweight language backbone, a Whisper-based audio encoder, and a sparsely activated Mixture-of-Experts (MoE) adapter that explicitly accounts for audio heterogeneity and alleviates cross-modal optimization conflicts under limited capacity. To further enhance paralinguistic reasoning, we introduce DataFlux, a closed loop audio instruction data synthesis and verification pipeline that constructs high quality, logically consistent supervision from raw audio. Extensive evaluations across ASR, knowledge reasoning, safety, instruction following, and paralinguistic benchmarks, demonstrate that Eureka-Audio achieves an efficient balance between computational cost and performance. These results establish Eureka Audio as a strong and practical baseline for lightweight audio understanding models.
title Eureka-Audio: Triggering Audio Intelligence in Compact Language Models
topic Sound
Artificial Intelligence
url https://arxiv.org/abs/2602.13954