{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/mellow-a-small-audio-language-model-for","title":"Mellow: a small audio language model for reasoning","arxiv_id":"2503.08540","date":"2025-03-11","proceeding":null,"authors":["Soham Deshmukh","Satvik Dixit","Rita Singh","Bhiksha Raj"],"abstract":"Multimodal Audio-Language Models (ALMs) can understand and reason over both audio and text. Typically, reasoning performance correlates with model size, with the best results achieved by models exceeding 8 billion parameters. However, no prior work has explored enabling small audio-language models to perform reasoning tasks, despite the potential applications for edge devices. To address this gap, we introduce Mellow, a small Audio-Language Model specifically designed for reasoning. Mellow achieves state-of-the-art performance among existing small audio-language models and surpasses several larger models in reasoning capabilities. For instance, Mellow scores 52.11 on MMAU, comparable to SoTA Qwen2 Audio (which scores 52.5) while using 50 times fewer parameters and being trained on 60 times less data (audio hrs). To train Mellow, we introduce ReasonAQA, a dataset designed to enhance audio-grounded reasoning in models. It consists of a mixture of existing datasets (30% of the data) and synthetically generated data (70%). The synthetic dataset is derived from audio captioning datasets, where Large Language Models (LLMs) generate detailed and multiple-choice questions focusing on audio events, objects, acoustic scenes, signal properties, semantics, and listener emotions. To evaluate Mellow's reasoning ability, we benchmark it on a diverse set of tasks, assessing on both in-distribution and out-of-distribution data, including audio understanding, deductive reasoning, and comparative reasoning. Finally, we conduct extensive ablation studies to explore the impact of projection layer choices, synthetic data generation methods, and language model pretraining on reasoning performance. Our training dataset, findings, and baseline pave the way for developing small ALMs capable of reasoning.","url_abs":"https://arxiv.org/abs/2503.08540v1","url_pdf":"https://arxiv.org/pdf/2503.08540v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"mellow-a-small-audio-language-model-for","repo_url":"https://github.com/soham97/mellow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"audio-captioning","task_name":"Audio captioning"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"multiple-choice","task_name":"Multiple-choice"},{"task_slug":"synthetic-data-generation","task_name":"Synthetic Data Generation"}],"methods":[{"method_slug":"set","method_name":"SET"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2503.08540","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2503.08540"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/soham97/mellow","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":2,"ran_fixture":1,"unverified":3},"by_repo_kind":{"official":{"samples":6,"ran":3,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"5337e68cdb46397d","entry":"do_mixup","repo":"soham97/mellow","repo_kind":"official","path":"mellow/model/htsat.py","file_url":"https://github.com/soham97/mellow/blob/HEAD/mellow/model/htsat.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5337e68cdb46397d"}},{"code_sha256_prefix":"3ac6b7d76e8e3584","entry":"drop_path","repo":"soham97/mellow","repo_kind":"official","path":"mellow/model/htsat.py","file_url":"https://github.com/soham97/mellow/blob/HEAD/mellow/model/htsat.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3ac6b7d76e8e3584"}},{"code_sha256_prefix":"bdcac6c19196f655","entry":"interpolate","repo":"soham97/mellow","repo_kind":"official","path":"mellow/model/htsat.py","file_url":"https://github.com/soham97/mellow/blob/HEAD/mellow/model/htsat.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"bdcac6c19196f655"}},{"code_sha256_prefix":"93883289ece4faef","entry":"downsample","repo":"soham97/mellow","repo_kind":"official","path":"mellow/model/decoder.py","file_url":"https://github.com/soham97/mellow/blob/HEAD/mellow/model/decoder.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"93883289ece4faef"}},{"code_sha256_prefix":"705be5651717fc89","entry":"get_audio_encoder","repo":"soham97/mellow","repo_kind":"official","path":"mellow/model/audio.py","file_url":"https://github.com/soham97/mellow/blob/HEAD/mellow/model/audio.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"705be5651717fc89"}},{"code_sha256_prefix":"3a331962cec26f5b","entry":"get_decoder","repo":"soham97/mellow","repo_kind":"official","path":"mellow/model/decoder.py","file_url":"https://github.com/soham97/mellow/blob/HEAD/mellow/model/decoder.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3a331962cec26f5b"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}