A Baseline for Self-state Identification and Classification in Mental Health Data: CLPsych 2025 Task
This work addresses the challenge of identifying adaptive or maladaptive self-states in mental health data, which is incremental as it builds on existing tasks and methods.
The paper tackled the problem of classifying self-states in mental health data from Reddit by developing a baseline system using few-shot learning with a quantized Gemma 2 9B model and a sentence-based preprocessing step, achieving a test-time recall of 0.579 and placing third out of fourteen submissions.
We present a baseline for the CLPsych 2025 A.1 task: classifying self-states in mental health data taken from Reddit. We use few-shot learning with a 4-bit quantized Gemma 2 9B model and a data preprocessing step which first identifies relevant sentences indicating self-state evidence, and then performs a binary classification to determine whether the sentence is evidence of an adaptive or maladaptive self-state. This system outperforms our other method which relies on an LLM to highlight spans of variable length independently. We attribute the performance of our model to the benefits of this sentence chunking step for two reasons: partitioning posts into sentences 1) broadly matches the granularity at which self-states were human-annotated and 2) simplifies the task for our language model to a binary classification problem. Our system places third out of fourteen systems submitted for Task A.1, achieving a test-time recall of 0.579.