{"paper":{"title":"Multimodal Ambivalence/Hesitancy Recognition in Videos for Personalized Digital Health Interventions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Standard deep learning models show limited success recognizing ambivalence and hesitancy in videos, indicating that better methods for handling multimodal conflicts are needed.","cross_cats":["cs.HC","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alessandro Lameiras Koerich, Eric Granger, Lorenzo Sia, Manuela Gonz\\'alez-Gonz\\'alez, Marco Pedersoli, Masoumeh Sharafi, Muhammad Haseeb Aslam, Muhammad Osama Zeeshan, Nicolas Richet, Simon L Bacon, Soufiane Belharbi","submitted_at":"2026-04-13T17:05:38Z","abstract_excerpt":"Using behavioural science, health interventions focus on behaviour change by providing a framework to help patients acquire and maintain healthy habits that improve medical outcomes. In-person interventions are costly and difficult to scale, especially in resource-limited regions. Digital health interventions offer a cost-effective approach, potentially supporting independent living and self-management. Automating such interventions, especially through machine learning, has recently gained considerable attention. Ambivalence and hesitancy (A/H) play a primary role for individuals to delay, avo"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our results show limited performance, suggesting that more adapted multi-modal models are required for accurate A/H recognition. Better methods for modeling spatio-temporal and multimodal fusion are necessary to leverage conflicts within/across modalities.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That standard deep learning architectures for video can capture subtle affective inconsistencies across and within modalities without major new adaptations for spatio-temporal fusion.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Multimodal deep learning for ambivalence/hesitancy recognition in videos yields limited results on the BAH dataset, highlighting the need for improved spatio-temporal and cross-modal fusion methods.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Standard deep learning models show limited success recognizing ambivalence and hesitancy in videos, indicating that better methods for handling multimodal conflicts are needed.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"986e596f1a587c237f369c7949e7ad6f8a12a04434657fc32bfd1330cc8af233"},"source":{"id":"2604.11730","kind":"arxiv","version":4},"verdict":{"id":"54ac16e9-2ebb-4196-af1c-71960c246511","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T15:56:49.245876Z","strongest_claim":"Our results show limited performance, suggesting that more adapted multi-modal models are required for accurate A/H recognition. Better methods for modeling spatio-temporal and multimodal fusion are necessary to leverage conflicts within/across modalities.","one_line_summary":"Multimodal deep learning for ambivalence/hesitancy recognition in videos yields limited results on the BAH dataset, highlighting the need for improved spatio-temporal and cross-modal fusion methods.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That standard deep learning architectures for video can capture subtle affective inconsistencies across and within modalities without major new adaptations for spatio-temporal fusion.","pith_extraction_headline":"Standard deep learning models show limited success recognizing ambivalence and hesitancy in videos, indicating that better methods for handling multimodal conflicts are needed."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.11730/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"d1b49c6209ccc2fb59020df247998ee36fef5a4d5f0c9454b423c66cb59c023e"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}