LLM embedding spaces partially recover expert-defined symptom structure in mental health language
A new arXiv preprint investigates whether LLM embedding geometry aligns with expert-defined symptom structure in mental health language, using 28 Reddit communities as a testbed. The authors compare pretrained and fine-tuned Qwen3 embeddings (0.6B and 4B) against an expert symptom matrix via representational similarity analysis, with controls for affective, stylistic, and topic confounds. Results show measurable but level-dependent alignment: fine-tuning strengthens it at fine-grained category levels, and larger scale improves both zero-shot alignment and fine-tuning gains. The paper argues that classification accuracy alone is insufficient to validate embedding geometry against domain knowledge.
Related guides (1)
Related events (8)
LLUMI: Fine-Tuning Open-Source LLMs for Mental Health Writing Assistance Using Reddit Community Feedback
LLUMI is a two-component system (a generation model and an improvement model) designed to provide mental health writing assistance using smaller open-source LLMs hosted in privacy-preserving, on-premise environments. The system leverages Reddit community endorsement signals (upvotes/downvotes) to construct preference pairs for SFT and DPO training, then further aligns outputs via human evaluation across readability, empathy, connection, actionability, and safety dimensions. Results show LLUMI achieves performance comparable to proprietary GPT-based models on linguistic and human evaluations, suggesting community-derived preference signals can substitute for expensive expert labeling in sensitive domains.
LLMs predict dementia and depression severity from clinical interview transcripts in zero-shot and feature-extraction settings
Researchers evaluate three open-weights LLMs (Mistral 3.1, DeepHermes, Qwen3) for predicting dementia and depression severity from speech transcripts of 154 German-speaking patients in standardized clinical interviews. The study introduces a new observer-based Global Depression Scale (GDS-D) and tests both zero-shot prediction and LLM-based feature extraction for Support Vector Regression. Zero-shot performs well for depression (MAE 0.60), while structured feature extraction reduces dementia assessment error by up to 35%; pause-enriched automatic transcripts match human transcription quality, suggesting viable fully-automated screening pipelines.
Fine-tuning LLMs to passively estimate depression severity from AI mental health conversations
Researchers fine-tune a Qwen3.5-27B model with a regression head to predict PHQ-9 depression severity scores directly from AI mental health app conversation transcripts, eliminating the need for explicit self-report completion. The training set of 6,283 users combines 3,111 ground-truth labels with pseudolabels generated by Claude Opus and iterative intermediate models. On a held-out test of 842 users, the best model achieves MAE=2.6, Pearson r=0.80, and AUC=0.91 at the clinical PHQ-9≥10 threshold, with AUC>0.87 across all severity thresholds. The work demonstrates a passive, continuous symptom-monitoring approach that could reduce response bias in mental health platforms.
Persona-conditioned LLM support for people who use drugs reveals tension between generic empathy and clinical alignment
Researchers present a proof-of-concept study using Latent Profile Analysis on Reddit data to identify four self-stigma personas among people who use drugs, then train classifiers to detect these personas from posting history (macro-F1 = 0.74 at 30 posts). Persona-matched LLM responses achieved targeted behavioral shifts, but clinical expert raters preferred the generic empathy of persona-neutral baselines. The core finding is a misalignment: holistic empathy judgments and clinically-aligned response design can pull in opposite directions, suggesting current evaluation rubrics for LLM-based mental health support are inadequate.
LLMs fail to consistently simulate demographic perspective-taking in hate speech annotation
A new arXiv paper evaluates whether persona-conditioned LLMs can replicate how different demographic groups perceive hate speech, testing three dimensions: inter-group disagreement, in-group sensitivity, and vicarious prediction. No model consistently captures all three dimensions, and performance is highly model-dependent rather than emerging reliably from identity prompts alone. Vicarious prompting with Llama 3.1 provides the closest approximation to human disagreement patterns across demographic axes. The findings have implications for using LLMs as proxies for diverse human annotators in content moderation tasks.
Automated ICD Classification of Psychiatric Diagnoses Using NLP and LLMs
This study evaluates NLP and ML approaches for automating the mapping of free-text psychiatric descriptions to ICD diagnostic codes, using a dataset of 145,513 Spanish clinical records. Methods range from classical BoW/TF-IDF representations to transformer-based embeddings including e5_large, BioLORD, and Llama-3-8B. Fine-tuned e5_large achieved the best performance with a micro-F1 of 0.866, outperforming classical methods by capturing semantic nuance and medical terminology. The work highlights challenges of long-tail label distributions and ambiguity specific to psychiatric clinical language.
Cross-lingual evaluation framework reveals LLMs redistribute cultural narrative structure while preserving semantic meaning
A new arXiv preprint introduces a multilingual evaluation framework using 414 proverbs across 15 languages to assess whether LLMs preserve culturally grounded meaning when generating narratives. Using four LLMs to produce 13k narratives, the study finds that cross-lingual prompting preserves proverb-level semantic meaning but systematically redistributes agency, social positioning, and narrative structure. Strong inter-model convergence across architectures suggests multilingual LLMs rely on shared semantic abstractions. The authors argue that semantic similarity metrics alone overestimate cultural preservation in multilingual evaluations.
Mechanistic analysis of how LLMs encode essay quality in internal representations
Researchers systematically probe the hidden representations of eight LLMs across three essay datasets (ASAP++, CSEE, ENEM) to understand how automated essay scoring (AES) works internally. Using linear probing, dimensionality reduction, and neuron-level analysis, they find essay quality is encoded in a linearly accessible form that emerges progressively across layers and partially transfers across prompts. Individual 'essay scoring neurons' are identified whose activations correlate with scores and respond to targeted interventions, with longer essays relying more on deeper layers. The work contributes to mechanistic interpretability of LLM-based scoring systems.