Baris Karacan

This study demonstrates that analyzing open-ended teacher narratives, using LLM-assisted theme discovery, can uncover distinct behavioral signals related to ADHD that are missed by traditional, structured rating scales.

The paper introduces a supervised fine-tuning pipeline using large language models to accurately categorize sentence-level clinical provenance across multi-disciplinary hospital notes, demonstrating that larger, quantized models are critical for effective cross-domain transfer.