SleepLM: Natural-Language Intelligence for Human Sleep
📖 Full Retelling
arXiv:2602.23605v1 Announce Type: new
Abstract: We present SleepLM, a family of sleep-language foundation models that enable human sleep alignment, interpretation, and interaction with natural language. Despite the critical role of sleep, learning-based sleep analysis systems operate in closed label spaces (e.g., predefined stages or events) and fail to describe, query, or generalize to novel sleep phenomena. SleepLM bridges natural language and multimodal polysomnography, enabling language-gro
Entity Intersection Graph
No entity connections available yet for this article.
Original Source
--> Computer Science > Artificial Intelligence arXiv:2602.23605 [Submitted on 27 Feb 2026] Title: SleepLM: Natural-Language Intelligence for Human Sleep Authors: Zongzhe Xu , Zitao Shuai , Eideen Mozaffari , Ravi S. Aysola , Rajesh Kumar , Yuzhe Yang View a PDF of the paper titled SleepLM: Natural-Language Intelligence for Human Sleep, by Zongzhe Xu and 5 other authors View PDF Abstract: We present SleepLM, a family of sleep-language foundation models that enable human sleep alignment, interpretation, and interaction with natural language. Despite the critical role of sleep, learning-based sleep analysis systems operate in closed label spaces (e.g., predefined stages or events) and fail to describe, query, or generalize to novel sleep phenomena. SleepLM bridges natural language and multimodal polysomnography, enabling language-grounded representations of sleep physiology. To support this alignment, we introduce a multilevel sleep caption generation pipeline that enables the curation of the first large-scale sleep-text dataset, comprising over 100K hours of data from more than 10,000 individuals. Furthermore, we present a unified pretraining objective that combines contrastive alignment, caption generation, and signal reconstruction to better capture physiological fidelity and cross-modal interactions. Extensive experiments on real-world sleep understanding tasks verify that SleepLM outperforms state-of-the-art in zero-shot and few-shot learning, cross-modal retrieval, and sleep captioning. Importantly, SleepLM also exhibits intriguing capabilities including language-guided event localization, targeted insight generation, and zero-shot generalization to unseen tasks. All code and data will be open-sourced. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2602.23605 [cs.AI] (or arXiv:2602.23605v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2602.23605 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: ...
Read full article at source