Academic Journal

Deploying Efficient LLM Agents on Maritime Autonomous Surface Ships: Fine-Tuning, RAG, and Function Calling in a Mid-Size Model.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Deploying Efficient LLM Agents on Maritime Autonomous Surface Ships: Fine-Tuning, RAG, and Function Calling in a Mid-Size Model.
Συγγραφείς: Ren, Yiling, Chen, Mozi, Weng, Junjie, Zhang, Shengkai, Xiao, Xuedou, Liu, Kezhong
Πηγή: Information; Mar2026, Vol. 17 Issue 3, p284, 15p
Θεματικοί όροι: Ships, Intelligent agents, Subroutines (Computer programs), Edge computing, Machine learning, Navigation, Natural language processing, Language models
Περίληψη: Deploying Large Language Models (LLMs) on Maritime Autonomous Surface Ships (MASS) entails a critical trade-off between reasoning depth, inference latency, and hardware constraints. To fill the existing gap, we introduce MARTIAN (Maritime Agent for Real-time Tactical Inference And Navigation), a 14B-parameter decision support agent engineered for edge deployment on standard vessel hardware (e.g., the NVIDIA Jetson AGX Orin). Central to our approach is the Cognitive Core architecture, which utilizes a verified dataset of 21,800 Chain-of-Thought (CoT) instruction–response pairs to align general linguistic capabilities with maritime procedural logic. Empirical evaluations demonstrate that MARTIAN achieves an overall accuracy of 73.23% (SFT only) and 81.16% (SFT + RAG) on the Bilingual Maritime Multiple-Choice Questionnaire (BM-MCQ), a standardized assessment dataset constructed based on Officer of the Watch (OOW) competencies. Notably, the SFT-only configuration attains 78.53% on pure-logic-intensive COLREG tasks—surpassing the 72B-parameter Qwen-2.5 foundation model in this domain—while maintaining a real-time inference latency of 22.4 ms/token. Crucially, our ablation studies support a nuanced Interference Hypothesis: while RAG significantly enhances factual recall in knowledge-intensive domains (boosting total accuracy from 73.23% to 81.16%), it concurrently introduces semantic noise that degrades performance in pure logic reasoning tasks (e.g., COLREG maneuvering accuracy decreases from 78.53% to 77.36%). On the basis of this finding, we identify and empirically motivate a decoupled cognitive design principle that separates procedural reflexes (via SFT) from declarative knowledge (via RAG). While the full implementation of an adaptive routing mechanism is deferred to future work, the ablation results presented herein offer a validated, cost-effective reference architecture for deploying transparent and regulation-compliant AI on resource-constrained merchant vessels. [ABSTRACT FROM AUTHOR]
Copyright of Information is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Βάση Δεδομένων: Complementary Index
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://resolver.ebsco.com/c/fiv2js/result?sid=EBSCO:edb&genre=article&issn=20782489&ISBN=&volume=17&issue=3&date=20260301&spage=284&pages=284-298&title=Information&atitle=Deploying%20Efficient%20LLM%20Agents%20on%20Maritime%20Autonomous%20Surface%20Ships%3A%20Fine-Tuning%2C%20RAG%2C%20and%20Function%20Calling%20in%20a%20Mid-Size%20Model.&aulast=Ren%2C%20Yiling&id=DOI:10.3390/info17030284
    Name: Full Text Finder (for New FTF UI) (ns324271)
    Category: fullText
    Text: Full Text Finder
    MouseOverText: Full Text Finder
Header DbId: edb
DbLabel: Complementary Index
An: 192595468
RelevancyScore: 1061
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 1060.7568359375
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Deploying Efficient LLM Agents on Maritime Autonomous Surface Ships: Fine-Tuning, RAG, and Function Calling in a Mid-Size Model.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Ren%2C+Yiling%22">Ren, Yiling</searchLink><br /><searchLink fieldCode="AR" term="%22Chen%2C+Mozi%22">Chen, Mozi</searchLink><br /><searchLink fieldCode="AR" term="%22Weng%2C+Junjie%22">Weng, Junjie</searchLink><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Shengkai%22">Zhang, Shengkai</searchLink><br /><searchLink fieldCode="AR" term="%22Xiao%2C+Xuedou%22">Xiao, Xuedou</searchLink><br /><searchLink fieldCode="AR" term="%22Liu%2C+Kezhong%22">Liu, Kezhong</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: Information; Mar2026, Vol. 17 Issue 3, p284, 15p
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Ships%22">Ships</searchLink><br /><searchLink fieldCode="DE" term="%22Intelligent+agents%22">Intelligent agents</searchLink><br /><searchLink fieldCode="DE" term="%22Subroutines+%28Computer+programs%29%22">Subroutines (Computer programs)</searchLink><br /><searchLink fieldCode="DE" term="%22Edge+computing%22">Edge computing</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br /><searchLink fieldCode="DE" term="%22Navigation%22">Navigation</searchLink><br /><searchLink fieldCode="DE" term="%22Natural+language+processing%22">Natural language processing</searchLink><br /><searchLink fieldCode="DE" term="%22Language+models%22">Language models</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Deploying Large Language Models (LLMs) on Maritime Autonomous Surface Ships (MASS) entails a critical trade-off between reasoning depth, inference latency, and hardware constraints. To fill the existing gap, we introduce MARTIAN (Maritime Agent for Real-time Tactical Inference And Navigation), a 14B-parameter decision support agent engineered for edge deployment on standard vessel hardware (e.g., the NVIDIA Jetson AGX Orin). Central to our approach is the Cognitive Core architecture, which utilizes a verified dataset of 21,800 Chain-of-Thought (CoT) instruction–response pairs to align general linguistic capabilities with maritime procedural logic. Empirical evaluations demonstrate that MARTIAN achieves an overall accuracy of 73.23% (SFT only) and 81.16% (SFT + RAG) on the Bilingual Maritime Multiple-Choice Questionnaire (BM-MCQ), a standardized assessment dataset constructed based on Officer of the Watch (OOW) competencies. Notably, the SFT-only configuration attains 78.53% on pure-logic-intensive COLREG tasks—surpassing the 72B-parameter Qwen-2.5 foundation model in this domain—while maintaining a real-time inference latency of 22.4 ms/token. Crucially, our ablation studies support a nuanced Interference Hypothesis: while RAG significantly enhances factual recall in knowledge-intensive domains (boosting total accuracy from 73.23% to 81.16%), it concurrently introduces semantic noise that degrades performance in pure logic reasoning tasks (e.g., COLREG maneuvering accuracy decreases from 78.53% to 77.36%). On the basis of this finding, we identify and empirically motivate a decoupled cognitive design principle that separates procedural reflexes (via SFT) from declarative knowledge (via RAG). While the full implementation of an adaptive routing mechanism is deferred to future work, the ablation results presented herein offer a validated, cost-effective reference architecture for deploying transparent and regulation-compliant AI on resource-constrained merchant vessels. [ABSTRACT FROM AUTHOR]
– Name: Abstract
  Label:
  Group: Ab
  Data: <i>Copyright of Information is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edb&AN=192595468
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.3390/info17030284
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 15
        StartPage: 284
    Subjects:
      – SubjectFull: Ships
        Type: general
      – SubjectFull: Intelligent agents
        Type: general
      – SubjectFull: Subroutines (Computer programs)
        Type: general
      – SubjectFull: Edge computing
        Type: general
      – SubjectFull: Machine learning
        Type: general
      – SubjectFull: Navigation
        Type: general
      – SubjectFull: Natural language processing
        Type: general
      – SubjectFull: Language models
        Type: general
    Titles:
      – TitleFull: Deploying Efficient LLM Agents on Maritime Autonomous Surface Ships: Fine-Tuning, RAG, and Function Calling in a Mid-Size Model.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Ren, Yiling
      – PersonEntity:
          Name:
            NameFull: Chen, Mozi
      – PersonEntity:
          Name:
            NameFull: Weng, Junjie
      – PersonEntity:
          Name:
            NameFull: Zhang, Shengkai
      – PersonEntity:
          Name:
            NameFull: Xiao, Xuedou
      – PersonEntity:
          Name:
            NameFull: Liu, Kezhong
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 03
              Text: Mar2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 20782489
          Numbering:
            – Type: volume
              Value: 17
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: Information
              Type: main
ResultId 1