Continuous glucose traces look like one noisy curve. They are not. They contain a slow baseline and short-term deviations from meals, activity, and sensor artifacts. GlucoFM is built around that split.

The model is a lightweight, self-supervised foundation model for continuous glucose monitoring. One stream tracks slower metabolic state. The other tracks transient events. Time of day is preserved. Missing readings are masked rather than invented. Pretraining predicts latent structure instead of reconstructing raw noise, and uses augmentations that mimic real device failures: drift, sparse sampling, brief disconnects.

A token is not a fixed unit of text. It is the output of a tokenizer. The same content can produce meaningfully different token counts depending on language, model version, and even quiet tokenizer updates. Non-English text routinely consumes more tokens. Output tokens are priced much higher than input tokens. Reasoning or “thinking” tokens are usually billed as output as well.

These mechanics create systematic opacity. Listed per-million-token prices are unreliable proxies for real cost. Cheaper models can end up more expensive overall because they consume more tokens to reach the same result. Subscription plans hide limits through silent model downgrades and non-comparable quotas. The practical consequence is that teams cannot budget accurately from published rate cards alone.

Wearable devices generate enormous amounts of multimodal sensor data (heart rate, motion, skin conductance, temperature, etc.), but most AI models for health are narrow, task-specific, and require large amounts of labeled data. This limits scalability and generalization in real-world settings where data is often fragmented or incomplete.

Google researchers introduce SensorFM, a large sensor foundation model pre-trained self-supervised on over one trillion minutes of wearable data from more than five million consented participants across diverse devices and geographies. It learns a general-purpose representation of human physiology using a missingness-aware framework (AIM) that treats real-world data gaps as meaningful signals rather than noise.

Single-cell RNA sequencing has generated massive datasets across tissues, experiments, and species, but integrating and analyzing them remains challenging due to batch effects, species differences, and the need for extensive annotations.

Researchers introduce UCE (Universal Cell Embedding), a foundation model trained self-supervised on 36 million cells. By representing cells as "bags of RNA" ordered by genomic location and leveraging protein language models (ESM2), UCE learns a unified latent space that captures biological variation while remaining robust to experimental noise.