{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,28]],"date-time":"2026-02-28T11:11:06Z","timestamp":1772277066242,"version":"3.50.1"},"reference-count":38,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,1,11]],"date-time":"2026-01-11T00:00:00Z","timestamp":1768089600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,1,11]],"date-time":"2026-01-11T00:00:00Z","timestamp":1768089600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100006190","name":"Research and Development","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006190","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,1,11]]},"DOI":"10.1109\/sii64115.2026.11404560","type":"proceedings-article","created":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T20:47:13Z","timestamp":1772225233000},"page":"794-799","source":"Crossref","is-referenced-by-count":0,"title":["EVLOD: Ensemble Vision-Language Open-vocabulary Detection for Construction Site Object Recognition"],"prefix":"10.1109","author":[{"given":"Yongdong","family":"Wang","sequence":"first","affiliation":[{"name":"The University of Tokyo,Tokyo,Japan,113-8656"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Runze","family":"Xiao","sequence":"additional","affiliation":[{"name":"The University of Tokyo,Tokyo,Japan,113-8656"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jun Younes Louhi","family":"Kasahara","sequence":"additional","affiliation":[{"name":"The University of Tokyo,Tokyo,Japan,113-8656"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shota","family":"Chikushi","sequence":"additional","affiliation":[{"name":"The University of Tokyo,Tokyo,Japan,113-8656"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Keiji","family":"Nagatani","sequence":"additional","affiliation":[{"name":"The University of Tokyo,Tokyo,Japan,113-8656"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Atsushi","family":"Yamashita","sequence":"additional","affiliation":[{"name":"The University of Tokyo,Tokyo,Japan,113-8656"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hajime","family":"Asama","sequence":"additional","affiliation":[{"name":"The University of Tokyo,Tokyo,Japan,113-8656"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Summary of the white paper on land, infrastructure, transport and tourism in japan, 2024","volume-title":"Ministry of Land, Infrastructure, Transport and Tourism, Japan, Tech. Rep.","year":"2024"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2015.07.022"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2020.103517"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.22260\/ISARC2018\/0102"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2017.08.031"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-8667.1995.tb00298.x"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1016\/j.compind.2021.103570"},{"key":"ref8","article-title":"Palm-e: An embodied multimodal language model","author":"Driess","year":"2023"},{"key":"ref9","first-page":"9118","article-title":"Language models as zero-shot planners: Extracting actionable knowledge for embodied agents","volume-title":"International conference on machine learning","author":"Huang"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/icra48891.2023.10161317"},{"key":"ref11","article-title":"Dart-llm: Dependency-aware multirobot task decomposition and execution using large language models","author":"Wang","year":"2024"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1016\/j.aei.2015.03.006"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2020.103116"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1016\/j.aei.2018.05.003"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2018.11.033"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2020.103085"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2019.102894"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2018.12.014"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2018.2876865"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-019-01247-4"},{"key":"ref21","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proceeding International Conference on Machine Learning","author":"Radford"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01069"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72970-6_3"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1016\/j.aei.2024.103075"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.22260\/ISARC2025\/0184"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2022.104499"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2023.105083"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2020.103482"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-45014-9_1"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-009-9124-7"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20077-9_21"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20080-9_42"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-022-01653-1"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19833-5_29"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2017.2738401"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00745"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA57147.2024.10611393"}],"event":{"name":"2026 IEEE\/SICE International Symposium on System Integration (SII)","location":"Cancun, Mexico","start":{"date-parts":[[2026,1,11]]},"end":{"date-parts":[[2026,1,14]]}},"container-title":["2026 IEEE\/SICE International Symposium on System Integration (SII)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11404435\/11404394\/11404560.pdf?arnumber=11404560","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,28]],"date-time":"2026-02-28T07:01:26Z","timestamp":1772262086000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11404560\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,11]]},"references-count":38,"URL":"https:\/\/doi.org\/10.1109\/sii64115.2026.11404560","relation":{},"subject":[],"published":{"date-parts":[[2026,1,11]]}}}