{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T19:41:12Z","timestamp":1788982872775,"version":"build-2803163510"},"reference-count":68,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/501100010951","name":"Drainage Services Department","doi-asserted-by":"publisher","award":["DEMP\/2023\/25"],"award-info":[{"award-number":["DEMP\/2023\/25"]}],"id":[{"id":"10.13039\/501100010951","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Automation in Construction"],"published-print":{"date-parts":[[2025,11]]},"DOI":"10.1016\/j.autcon.2025.106490","type":"journal-article","created":{"date-parts":[[2025,8,29]],"date-time":"2025-08-29T14:14:58Z","timestamp":1756476898000},"page":"106490","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":21,"special_numbering":"C","title":["Enhancing visual-LLM for construction site safety compliance via prompt engineering and Bi-stage retrieval-augmented generation"],"prefix":"10.1016","volume":"179","author":[{"given":"Koi Xiaowen","family":"Guo","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peter Kok-Yiu","family":"Wong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jack C.P.","family":"Cheng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chak-Fu","family":"Chan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pak-Him","family":"Leung","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xingyu","family":"Tao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.autcon.2025.106490_bb0005","doi-asserted-by":"crossref","DOI":"10.1061\/(ASCE)CO.1943-7862.0002337","article-title":"Empirical Comparison of Learning Effectiveness of Immersive Virtual Reality\u2013Based Safety Training for Novice and Experienced Construction Workers","volume":"148","author":"Yu","year":"2022","journal-title":"J. Constr. Eng. Manag."},{"key":"10.1016\/j.autcon.2025.106490_bb0010","author":"U.S. Department of Labor"},{"key":"10.1016\/j.autcon.2025.106490_bb0015","author":"Eruostat"},{"key":"10.1016\/j.autcon.2025.106490_bb0020","author":"Labour Department"},{"key":"10.1016\/j.autcon.2025.106490_bb0025","doi-asserted-by":"crossref","DOI":"10.1016\/j.autcon.2022.104312","article-title":"Leung, vision-based monitoring of site safety compliance based on worker re-identification and personal protective equipment classification","volume":"139","author":"Cheng","year":"2022","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0030","doi-asserted-by":"crossref","DOI":"10.1016\/j.autcon.2021.103940","article-title":"Computer vision applications in construction: current state, opportunities & challenges","volume":"132","author":"Paneru","year":"2021","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0035","series-title":"International Symposium on Automation and Robotics in Construction (ISARC) Proceedings","first-page":"399","article-title":"Predicting Safety Hazards Among Construction Workers and Equipment Using Computer Vision and Deep Learning Techniques","author":"Wang","year":"2019"},{"key":"10.1016\/j.autcon.2025.106490_bb0040","doi-asserted-by":"crossref","DOI":"10.1016\/j.autcon.2019.103016","article-title":"Full body pose estimation of construction equipment using computer vision and deep learning techniques","volume":"110","author":"Luo","year":"2020","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0045","doi-asserted-by":"crossref","DOI":"10.1016\/j.autcon.2020.103444","article-title":"Construction machine pose prediction considering historical motions and activity attributes using gated recurrent unit (GRU)","volume":"121","author":"Luo","year":"2021","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0050","doi-asserted-by":"crossref","DOI":"10.1016\/j.compind.2021.103448","article-title":"Utilizing safety rule correlation for mobile scaffolds monitoring leveraging deep convolution neural networks","volume":"129","author":"Khan","year":"2021","journal-title":"Comput. Ind."},{"key":"10.1016\/j.autcon.2025.106490_bb0055","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1016\/j.autcon.2018.02.018","article-title":"Falls from heights: a computer vision-based approach for safety harness detection","volume":"91","author":"Fang","year":"2018","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0060","doi-asserted-by":"crossref","DOI":"10.1016\/j.autcon.2022.104670","article-title":"Intelligent question answering method for construction safety hazard knowledge based on deep semantic mining","volume":"145","author":"Tian","year":"2023","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0065","doi-asserted-by":"crossref","DOI":"10.1016\/j.autcon.2022.104535","article-title":"Automatic construction site hazard identification integrating construction scene graphs with BERT based domain knowledge","volume":"142","author":"Zhang","year":"2022","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0070","doi-asserted-by":"crossref","DOI":"10.1016\/j.autcon.2023.105158","article-title":"Augmented reality, deep learning and vision-language query system for construction worker safety","volume":"157","author":"Chen","year":"2024","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0075","article-title":"Training language models to follow instructions with human feedback","author":"Ouyang","year":"2022","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0080","article-title":"Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedback","author":"Peng","year":"2023","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0085","article-title":"Finetuned Language Models Are Zero-Shot Learners","author":"Wei","year":"2022","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0090","article-title":"A survey on hallucination in large language models: principles, taxonomy, challenges, and open questions","author":"Huang","year":"2024","journal-title":"ACM Trans. Inf. Syst."},{"key":"10.1016\/j.autcon.2025.106490_bb0095","article-title":"SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training","author":"Chu","year":"2025","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0100","article-title":"An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning","author":"Luo","year":"2025","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0105","article-title":"Reason-RFT: Reinforcement Fine-Tuning for Visual Reasoning","author":"Tan","year":"2025","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0110","article-title":"Precise Zero-Shot Dense Retrieval without Relevance Labels","author":"Gao","year":"2022","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0115","article-title":"Query Rewriting for Retrieval-Augmented Large Language Models","author":"Ma","year":"2023","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0120","series-title":"C&SD\u202f: Wages and Labour Earnings, Wages and Labour Earnings","author":"Census and Statistics Department","year":"2025"},{"key":"10.1016\/j.autcon.2025.106490_bb0125","doi-asserted-by":"crossref","DOI":"10.1016\/j.aei.2020.101060","article-title":"Automated text classification of near-misses from safety reports: an improved deep learning approach","volume":"44","author":"Fang","year":"2020","journal-title":"Adv. Eng. Inform."},{"key":"10.1016\/j.autcon.2025.106490_bb0130","doi-asserted-by":"crossref","DOI":"10.1016\/j.autcon.2022.104465","article-title":"Automated detection of contractual risk clauses from construction specifications using bidirectional encoder representations from transformers (BERT)","volume":"142","author":"Moon","year":"2022","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0135","doi-asserted-by":"crossref","DOI":"10.1016\/j.autcon.2023.104951","article-title":"Generating risk response measures for subway construction by fusion of knowledge and deep learning","volume":"152","author":"Zhou","year":"2023","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0140","doi-asserted-by":"crossref","DOI":"10.1016\/j.autcon.2022.104317","article-title":"Improving knowledge capture and retrieval in the BIM environment: combining case-based reasoning and natural language processing","volume":"139","author":"Wang","year":"2022","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0145","author":"OpenAI"},{"key":"10.1016\/j.autcon.2025.106490_bb0150","article-title":"InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks","author":"Chen","year":"2024","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0155","article-title":"CogVLM: Visual Expert for Pretrained Language Models","author":"Wang","year":"2024","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0160","article-title":"A Survey of Large Language Models","author":"Zhao","year":"2024","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0165","doi-asserted-by":"crossref","DOI":"10.1016\/j.autcon.2024.105542","article-title":"Recognizing temporary construction site objects using CLIP-based few-shot learning and multi-modal prototypes","volume":"165","author":"Liang","year":"2024","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0170","doi-asserted-by":"crossref","DOI":"10.1016\/j.eswa.2024.124601","article-title":"AutoRepo: a general framework for multimodal LLM-based automated construction reporting","volume":"255","author":"Pu","year":"2024","journal-title":"Expert Syst. Appl."},{"key":"10.1016\/j.autcon.2025.106490_bb0175","doi-asserted-by":"crossref","first-page":"625","DOI":"10.1038\/s41586-024-07421-0","article-title":"Detecting hallucinations in large language models using semantic entropy","volume":"630","author":"Farquhar","year":"2024","journal-title":"Nature"},{"key":"10.1016\/j.autcon.2025.106490_bb0180","series-title":"Lille, France","article-title":"Assisting in the identification of ergonomic risks for workers: a large vision-language model approach","author":"Fan","year":"2024"},{"key":"10.1016\/j.autcon.2025.106490_bb0185","article-title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","author":"Han","year":"2024","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0190","article-title":"Parameter-Efficient Fine-Tuning for Foundation Models","author":"Zhang","year":"2025","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0195","article-title":"An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale","author":"Dosovitskiy","year":"2020","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0200","article-title":"How do large language models capture the ever-changing world knowledge? A review of recent","volume":"arXiv","author":"Zhang","year":"2023","journal-title":"Advances"},{"key":"10.1016\/j.autcon.2025.106490_bb0205","article-title":"Rethinking with Retrieval: Faithful Large Language Model Inference","author":"He","year":"2022","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0210","article-title":"Self-Consistency Improves Chain of Thought Reasoning in Language Models","author":"Wang","year":"2022","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0215","first-page":"24824","article-title":"Chain-of-thought prompting elicits reasoning in large language models","volume":"35","author":"Wei","year":"2022","journal-title":"Adv. Neural Inf. Proces. Syst."},{"key":"10.1016\/j.autcon.2025.106490_bb0220","author":"Gao"},{"key":"10.1016\/j.autcon.2025.106490_bb0225","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3439726","article-title":"Deep Learning--based Text Classification: A Comprehensive Review","volume":"54","author":"Minaee","year":"2022","journal-title":"ACM Comput. Surv."},{"key":"10.1016\/j.autcon.2025.106490_bb0230","doi-asserted-by":"crossref","DOI":"10.1016\/j.aei.2025.103158","article-title":"Retrieval augmented generation-driven information retrieval and question answering in construction management","volume":"65","author":"Wu","year":"2025","journal-title":"Adv. Eng. Inform."},{"key":"10.1016\/j.autcon.2025.106490_bb0235","article-title":"Construction contract risk identification based on knowledge-augmented language models","volume":"157\u2013158","author":"Wong","year":"2024","journal-title":"Comput. Ind."},{"key":"10.1016\/j.autcon.2025.106490_bb0240","author":"Inc. NetEase Youdao"},{"key":"10.1016\/j.autcon.2025.106490_bb0245","article-title":"Take a Step Back: Evoking Reasoning via Abstraction in Large Language Models","author":"Zheng","year":"2024","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0250","author":"nlmatics"},{"key":"10.1016\/j.autcon.2025.106490_bb0255","author":"nlmatics"},{"key":"10.1016\/j.autcon.2025.106490_bb0260","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1016\/j.autcon.2012.11.037","article-title":"Content-based text mining technique for retrieval of CAD documents","volume":"31","author":"Yu","year":"2013","journal-title":"Autom. Constr."},{"key":"10.1016\/j.autcon.2025.106490_bb0265","series-title":"Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)","first-page":"3982","article-title":"Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks","author":"Reimers","year":"2019"},{"key":"10.1016\/j.autcon.2025.106490_bb0270","author":"Isom"},{"key":"10.1016\/j.autcon.2025.106490_bb0275","series-title":"The Faiss library","volume":"arXiv","author":"Douze","year":"2024"},{"key":"10.1016\/j.autcon.2025.106490_bb0280","doi-asserted-by":"crossref","first-page":"824","DOI":"10.1109\/TPAMI.2018.2889473","article-title":"Efficient and robust approximate nearest neighbor search using hierarchical navigable small world graphs","volume":"42","author":"Malkov","year":"2020","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"10.1016\/j.autcon.2025.106490_bb0285","article-title":"From word vectors to multimodal Embeddings: techniques","volume":"arXiv","author":"Zhang","year":"2024","journal-title":"Applications, and Future Directions For Large Language Models"},{"key":"10.1016\/j.autcon.2025.106490_bb0290","series-title":"Improvements to BM25 and Language Models Examined, in: Proceedings of the 19th Australasian Document Computing Symposium","first-page":"58","author":"Trotman","year":"2014"},{"key":"10.1016\/j.autcon.2025.106490_bb0295","author":"Youdao"},{"key":"10.1016\/j.autcon.2025.106490_bb0300","article-title":"RAG-Gym: Optimizing Reasoning and Search Agents with Process Supervision","author":"Xiong","year":"2025","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0305","article-title":"Least-to-Most Prompting Enables Complex Reasoning in Large Language Models","author":"Zhou","year":"2022","journal-title":"arXiv"},{"key":"10.1016\/j.autcon.2025.106490_bb0310","doi-asserted-by":"crossref","first-page":"95","DOI":"10.1007\/s10648-023-09817-2","article-title":"The development of cognitive load theory: replication crises and incorporation of other theories can Lead to theory expansion","volume":"35","author":"Sweller","year":"2023","journal-title":"Educ. Psychol. Rev."},{"key":"10.1016\/j.autcon.2025.106490_bb0315","doi-asserted-by":"crossref","first-page":"2348","DOI":"10.3758\/s13414-020-01997-1","article-title":"Global precedence changes by environment: a systematic review and meta-analysis on effect of perceptual field variables on global-local visual processing","volume":"82","author":"Rezvani","year":"2020","journal-title":"Atten. Percept. Psychophysiol."},{"key":"10.1016\/j.autcon.2025.106490_bb0320","doi-asserted-by":"crossref","first-page":"643","DOI":"10.1037\/0033-2909.129.5.643","article-title":"Visual feature integration theory: past, present, and future","volume":"129","author":"Quinlan","year":"2003","journal-title":"Psychol. Bull."},{"key":"10.1016\/j.autcon.2025.106490_bb0325","series-title":"Computer Vision \u2013 ECCV 2016","first-page":"852","article-title":"Visual relationship detection with language priors","author":"Lu","year":"2016"},{"key":"10.1016\/j.autcon.2025.106490_bb0330","doi-asserted-by":"crossref","first-page":"32","DOI":"10.1007\/s11263-016-0981-7","article-title":"Visual genome: connecting language and vision using crowdsourced dense image annotations","volume":"123","author":"Krishna","year":"2017","journal-title":"Int. J. Comput. Vis."},{"key":"10.1016\/j.autcon.2025.106490_bb0335","author":"Brown"},{"key":"10.1016\/j.autcon.2025.106490_bb0340","article-title":"A Survey on Hallucination in Large Vision-Language Models","author":"Liu","year":"2024","journal-title":"arXiv"}],"container-title":["Automation in Construction"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0926580525005308?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0926580525005308?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2025,10,30]],"date-time":"2025-10-30T01:21:49Z","timestamp":1761787309000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0926580525005308"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11]]},"references-count":68,"alternative-id":["S0926580525005308"],"URL":"https:\/\/doi.org\/10.1016\/j.autcon.2025.106490","relation":{},"ISSN":["0926-5805"],"issn-type":[{"value":"0926-5805","type":"print"}],"subject":[],"published":{"date-parts":[[2025,11]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Enhancing visual-LLM for construction site safety compliance via prompt engineering and Bi-stage retrieval-augmented generation","name":"articletitle","label":"Article Title"},{"value":"Automation in Construction","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.autcon.2025.106490","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}],"article-number":"106490"}}