{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T16:44:24Z","timestamp":1777653864782,"version":"3.51.4"},"reference-count":40,"publisher":"IEEE","license":[{"start":{"date-parts":[[2024,4,14]],"date-time":"2024-04-14T00:00:00Z","timestamp":1713052800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,4,14]],"date-time":"2024-04-14T00:00:00Z","timestamp":1713052800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2024,4,14]]},"DOI":"10.1109\/icasspw62465.2024.10627466","type":"proceedings-article","created":{"date-parts":[[2024,8,15]],"date-time":"2024-08-15T17:19:18Z","timestamp":1723742358000},"page":"655-659","source":"Crossref","is-referenced-by-count":6,"title":["Multimodal Attention Merging for Improved Speech Recognition and Audio Event Classification"],"prefix":"10.1109","author":[{"given":"Anirudh S.","family":"Sundar","sequence":"first","affiliation":[{"name":"Georgia Institute of Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chao-Han Huck","family":"Yang","sequence":"additional","affiliation":[{"name":"Amazon Alexa AI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"David M.","family":"Chan","sequence":"additional","affiliation":[{"name":"Amazon Alexa AI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shalini","family":"Ghosh","sequence":"additional","affiliation":[{"name":"Amazon Alexa AI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Venkatesh","family":"Ravichandran","sequence":"additional","affiliation":[{"name":"Amazon Alexa AI"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Phani Sankar","family":"Nidadavolu","sequence":"additional","affiliation":[{"name":"Amazon Alexa AI"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Language models are few-shot learners","volume-title":"NeurIPS","author":"Brown"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3122291"},{"key":"ref3","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2020"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i7.20729"},{"key":"ref5","article-title":"Voice2series: Reprogramming acoustic models for time series classification","volume-title":"ICML","author":"Yang"},{"key":"ref6","article-title":"I can\u2019t believe there\u2019s no images! learning visual tasks using only language data","author":"Gu","year":"2022"},{"key":"ref7","article-title":"Can wikipedia help offline reinforcement learning?","author":"Reid","year":"2022"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-long.504"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW60793.2023.00308"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.23915\/distill.00030"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i10.21295"},{"key":"ref13","author":"Devlin","year":"2019","journal-title":"Bert: Pre-training of deep bidirectional transformers for language understanding"},{"key":"ref14","article-title":"Beats: Audio pre-training with acoustic tokenizers","author":"Chen","year":"2022"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.105"},{"key":"ref16","article-title":"lo-fi: distributed fine-tuning without communication","author":"Wortsman","year":"2022"},{"key":"ref17","article-title":"Fusing finetuned models for better pretraining","author":"Choshen","year":"2022"},{"key":"ref18","article-title":"Model fusion via optimal transport","volume-title":"NeurIPS","author":"Singh"},{"key":"ref19","article-title":"Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time","volume-title":"ICML","author":"Wortsman"},{"key":"ref20","article-title":"Generalized sliced wasserstein distances","volume-title":"NeurIPS","author":"Kolouri"},{"key":"ref21","article-title":"Otkge: Multi-modal knowledge graph embeddings via optimal transport","volume-title":"NeurIPS","author":"Cao"},{"key":"ref22","article-title":"Merging models with fisher-weighted averaging","volume-title":"NeurIPS","author":"Matena"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/WACV45572.2020.9093365"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-eacl.153"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-10928-8_50"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.279"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref28","article-title":"Editing models with task arithmetic","author":"Ilharco","year":"2022"},{"issue":"1","key":"ref29","first-page":"2930","article-title":"Are all layers created equal?","volume":"23","author":"Zhang","year":"2022","journal-title":"JMLR"},{"key":"ref30","first-page":"7669","article-title":"Libri-light: A benchmark for asr with limited or no supervision","volume-title":"ICASSP 2020","author":"Kahn"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref32","volume-title":"The lj speech dataset","author":"Ito","year":"2017"},{"key":"ref33","author":"Yamagishi","year":"2019","journal-title":"Cstr vctk corpus: English multi-speaker corpus for cstr voice cloning toolkit (version 0.92)"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7952261"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/2733373.2806390"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-1079"},{"key":"ref38","article-title":"Decoupled weight decay regularization","volume-title":"ICLR","author":"Loshchilov"},{"key":"ref39","article-title":"Parameter-efficient transfer learning for nlp","volume-title":"ICML","author":"Houlsby"},{"key":"ref40","article-title":"Lora: Low-rank adaptation of large language models","volume-title":"ICLR","author":"J Hu"}],"event":{"name":"2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)","location":"Seoul, Korea, Republic of","start":{"date-parts":[[2024,4,14]]},"end":{"date-parts":[[2024,4,19]]}},"container-title":["2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10625769\/10625780\/10627466.pdf?arnumber=10627466","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,16]],"date-time":"2024-08-16T05:32:19Z","timestamp":1723786339000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10627466\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,4,14]]},"references-count":40,"URL":"https:\/\/doi.org\/10.1109\/icasspw62465.2024.10627466","relation":{},"subject":[],"published":{"date-parts":[[2024,4,14]]}}}