{"ID":2868658,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.15680","arxiv_id":"2509.15680","title":"SAM: A Mamba-2 State-Space Audio-Language Model","abstract":"We present SAM, a State-space Audio-language Model that integrates an audio encoder with a Mamba-2 backbone. SAM-2.7B achieves 21.1 mAP on AudioSet and 17.6 SPICE on AudioCaps, matching or surpassing larger 7B transformer-based models with fewer parameters. We further provide the first systematic, representation-level analysis of how SSMs interact with audio encoder outputs: (1) joint audio encoder finetuning is essential, supported by accuracy gains and observed adaptation of token representation rank and similarity across different SSM sizes; (2) despite linear scaling, SSMs benefit more from compact, information-rich audio token representations than from excessively long token sequences; and (3) incorporating instruction-following supervision substantially improves reasoning ability, boosting MMAU-Sound accuracy from 22.8 to 56.8. Through comprehensive experiments and analysis, we establish practical design principles for SSMs as strong, scalable backbones for audio-language models.","short_abstract":"We present SAM, a State-space Audio-language Model that integrates an audio encoder with a Mamba-2 backbone. SAM-2.7B achieves 21.1 mAP on AudioSet and 17.6 SPICE on AudioCaps, matching or surpassing larger 7B transformer-based models with fewer parameters. We further provide the first systematic, representation-level...","url_abs":"https://arxiv.org/abs/2509.15680","url_pdf":"https://arxiv.org/pdf/2509.15680v2","authors":"[\"Taehan Lee\",\"Jaehan Jung\",\"Hyukjun Lee\"]","published":"2025-09-19T06:54:42Z","proceeding":"cs.SD","tasks":"[\"cs.SD\",\"eess.AS\"]","methods":"[\"Transformer\",\"Language Model\"]","has_code":false}