Forgotten Words: Benchmarking NeoBERT for Dementia Detection in Low-Resource Conversational Filipino and English Speech
arXiv:2605.26007v1 Announce Type: new Abstract: Dementia detection from spontaneous speech offers a scalable approach to cognitive screening, yet NLP systems remain predominantly English-centric. This limitation is especially acute in the Philippines, where Filipino-English code-switching is pervasive and no prior work has addressed NLP-based dementia detection. We present the first systematic evaluation of transformer-based dementia detection in Filipino speech and the first assessment of NeoBERT in a clinical NLP setting. To separate language from domain effects, we construct a parallel bili
This research is happening now due to growing interest in applying AI for healthcare diagnostics and the increasing awareness of NLP limitations in linguistically diverse, low-resource contexts.
A strategic reader should care because it demonstrates the expansion of AI's diagnostic capabilities into new linguistic and cultural domains, with implications for global health and AI's societal impact.
The development of reliable dementia detection tools in languages like Filipino and English code-switching introduces new avenues for early diagnosis and personalized care in underserved regions.
- · Healthcare providers in linguistically diverse regions
- · NLP researchers
- · Patients with neurodegenerative diseases
- · AI-driven diagnostic companies
- · Traditional diagnostic methods
- · English-only AI solution providers
Improved early detection of dementia in low-resource language communities.
Increased demand for culturally and linguistically appropriate AI healthcare solutions, fostering local AI development.
Potential for AI to reduce healthcare disparities and improve health equity globally.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL