Evidence-Grounded Guardrail Extraction for Activity Recognition in Smart Homes using Small Language Models

• Item type: SIG Technical Reports(1)
• 公開日: 2026-05-06
• タイトル / 言語: ja
• タイトル: Evidence-Grounded Guardrail Extraction for Activity Recognition in Smart Homes using Small Language Models
• タイトル / 言語: en
• タイトル: Evidence-Grounded Guardrail Extraction for Activity Recognition in Smart Homes using Small Language Models
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: MBL
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: Nara Institute of Science and Technology／University of the Philippines Tacloban College
• 著者所属: Nara Institute of Science and Technology
• 著者所属: Okayama University／Nara Institute of Science and Technology
• 著者所属: Nara Institute of Science and Technology
• 著者所属: Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology / University of the Philippines Tacloban College
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Okayama University / Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者名: Victor,Romero II; Tomokazu,Matsui; Yuki,Matsuda; Hirohiko,Suwa; Keiichi,Yasumoto
• 著者名(英): Victor Romero II; Tomokazu Matsui; Yuki Matsuda; Hirohiko Suwa; Keiichi Yasumoto
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: Activity recognition remains a significant challenge in pervasive computing, where models must infer user actions from sparse signals but often fail to enforce the contextual constraints required for consistent predictions. This study proposes a failure-triggered method for extracting decision guardrails using Small Language Models (SLMs) to improve the reliability of activity recognition systems. The framework constructs a Knowledge Graph of guardrails from model feedback, which serves as a grounded evidence base for subsequent inference. During inference, these grounded constraints are incorporated to guide predictions toward more contextually consistent activity recognition. We implement a pipeline that generates and reuses these constraints, and evaluate its impact on classification performance and reasoning behaviour. Experiments show that the proposed approach improves top-1 accuracy from 46.4% to 69.7%, with reduced class-level confusion and more consistent predictions. This work offers a training-free mechanism for transitioning from purely pattern-based activity recognition toward more constraint-aware systems.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: Activity recognition remains a significant challenge in pervasive computing, where models must infer user actions from sparse signals but often fail to enforce the contextual constraints required for consistent predictions. This study proposes a failure-triggered method for extracting decision guardrails using Small Language Models (SLMs) to improve the reliability of activity recognition systems. The framework constructs a Knowledge Graph of guardrails from model feedback, which serves as a grounded evidence base for subsequent inference. During inference, these grounded constraints are incorporated to guide predictions toward more contextually consistent activity recognition. We implement a pipeline that generates and reuses these constraints, and evaluate its impact on classification performance and reasoning behaviour. Experiments show that the proposed approach improves top-1 accuracy from 46.4% to 69.7%, with reduced class-level confusion and more consistent predictions. This work offers a training-free mechanism for transitioning from purely pattern-based activity recognition toward more constraint-aware systems.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AA11851388
• 書誌情報: 研究報告モバイルコンピューティングと新社会システム（MBL）; 巻 2026-MBL-119, 号 31, p. 1-8, 発行日 2026-05-06
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8817
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会