The Flow has not summarised this story yet — read it at arXiv Machine Learning.
The study investigates how a large language model, Gemma-3-27B-PT, internally represents depressive symptoms. By applying mechanistic interpretability methods to the model’s residual stream, researchers found that symptom groups are geometrically distinct at layer 21, and that projected symptom vectors align with clinician-annotated rankings across mood, somatic, and suicidality dimensions. Additionally, a single depression vector at this layer can differentiate depressive from non-depressive text with an AUC of 0.789, suggesting a potential emotional valence gate for symptom projection.
arXiv:2607. 21496v1 Announce Type: cross Abstract: Cognitive impairment (CI) is a growing public health concern.
arXiv:2606. 19640v1 Announce Type: cross Abstract: AI and large language models (LLMs) have emerged as promising tools to address global mental health challenges.
arXiv:2609.01833v1 Announce Type: new Abstract: Sentence-level recognition of depression symptoms is challenging because similar expressions can differ in symptom relevance, and language-model infere...
arXiv:2608.30857v1 Announce Type: cross Abstract: Depression and anxiety in older adults with Mild Cognitive Impairment (MCI) are frequently underdiagnosed due to limited access to care. Multimodal a...
arXiv:2608.29582v1 Announce Type: cross Abstract: Current evaluations of large language models (LLMs) primarily focus on factual knowledge retrieval, overlooking the fundamental challenge of navigati...