Agentic data science with the Data Agent Kit
cloud.google.com·6h ago
AI Research
arXiv CS.AI·Aug 15·5 min read
AnchorSIPS: A Synthetic Dataset and Evaluation Resource for Evidence-Supported Psychosis-Risk Symptom Measurement
TL;DR
AnchorSIPS is a synthetic dataset of 10,000 structured psychosis-risk interviews designed to overcome data-sharing limitations in psychosis-risk assessment. It enables researchers to evaluate evidence-supported symptom measurement while maintaining patient privacy.
✦ Why It Matters
Explore AnchorSIPS to enhance your research on psychosis-risk assessment and evidence extraction.
Key Takeaways
- Utilize AnchorSIPS for psychosis-risk assessment research.
- Leverage structured interviews to improve evidence extraction methods.
- Explore the limitations of current LLMs in extracting follow-up details.
- Consider patient privacy when designing clinical data-sharing solutions.
How It Works
AnchorSIPS employs a plan-then-realize pipeline where a hidden case sheet defines the patient's clinical state. A deterministic planner structures the interview, and a large language model generates patient responses, ensuring consistency across dialogue turns.