Newest papers

NarraSpace: Multimodal Scene-Grounded Evidence for Picture-Description Dementia Classification

Qingming Li, Ruiyan Sun, Satoshi Nakamura · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28616 · cited by 0 works (OpenAlex) · open-access PDF

Korean

NarraSpace는 단어와 음성 시간을 이름화된 이미지 요소에 연결하고 미리 정의된 장면 이벤트에 대한 발언을 검색합니다. 설명 공간 설명기는 커버 및 시간 조직을 요약합니다. NarraSpace는 알츠하이머 질병 및 건강 통제 훈련 참가자의 예를 들어 각 사건의 텍스트를 냉동 언어 모델과 대조합니다. 저장된 소스 레코드가 소스 발언, 연설 간격 및 논리적 일치에 이벤트 점수를 연결합니다. 알츠하이머의 Dementia 인식의 공식 48 참가자 테스트 세트에서 자발적인 언어 벤치마크를 통해 각 사건의 텍스트를 대조하면 고정 지원 벡터 분류는 원본 문자 분류에 의해 설명할 수있는 $80%\2.56%의 평균 정밀

Machine translation (Meta M2M100); the English original follows.

Abstract

Picture-description dementia classification uses a shared visual scene, but a transcript-level prediction alone does not identify the scene evidence underlying its inputs. NarraSpace links words and speech times to named picture elements and retrieves utterances for predefined scene events. Narrative-spatial descriptors summarize coverage and temporal organization. A frozen language model contrasts each event's text under examples from Alzheimer's disease and healthy-control training participants, alongside transcript surprisal and coverage measurements. Retained source records link event scores to their source utterances, speech intervals, and lexical matches. On the official 48-participant test set of the Alzheimer's Dementia Recognition through Spontaneous Speech benchmark, a fixed support vector classifier achieves $80.88\%\pm2.56\%$ mean accuracy over fifty seeds; averaged probabilities yield $81.25\%$ accuracy and $91.32\%$ area under the receiver operating characteristic curve. NarraSpace makes the construction of scene-linked classification inputs explicit and inspectable by connecting event-indexed language-model measurements to their source utterances, speech timing, and extraction records. Such source-linked measurements could support clinician review of picture descriptions by presenting quantitative summaries alongside the original speech evidence.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Classifier (UML), Dementia, Natural language processing, Speech recognition, Artificial intelligence

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange