Authors share a new dataset of GPT-4 behavior-change conversations with user language metrics, perception measures, and feedback collected in a preregistered study.
Title resolution pending
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
representative citing papers
LLMs produce interpretive closure in 87.5% of ambiguous social scenarios through narrative alignment, reversal, or normative advice, with first-person perspectives increasing alignment tendencies.
SynDocDis generates synthetic physician-to-physician dialogues from metadata using LLMs and achieves high physician-rated quality in oncology and hepatology scenarios.
Warmth and cognitive empathy in LLMs drive higher anthropomorphism, trust, and relational closeness, especially on personal topics, while competence affects usefulness but not perceived human-likeness.
RECAP is an inference-time framework using cognitive appraisal theory to enhance emotional alignment and transparency in medical dialogue systems across model scales.
Proposes a multi-layer framework and agent architecture that operationalizes adaptation, coherence, continuity, and agency for longitudinal health AI agents.
A pilot study of 29 physicians found high perceived time savings (4.27/5) and decision support (4.16/5) from DR. INFO, yielding an NPS of 81.2 that dropped to 44.8 under conservative assumptions.
citing papers explorer
-
"You tell me": A Dataset of GPT-4-Based Behaviour Change Support Conversations
Authors share a new dataset of GPT-4 behavior-change conversations with user language metrics, perception measures, and feedback collected in a preregistered study.
-
What Did They Mean? How LLMs Resolve Ambiguous Social Situations across Perspectives and Roles
LLMs produce interpretive closure in 87.5% of ambiguous social scenarios through narrative alignment, reversal, or normative advice, with first-person perspectives increasing alignment tendencies.
-
SynDocDis: A Metadata-Driven Framework for Generating Synthetic Physician Discussions Using Large Language Models
SynDocDis generates synthetic physician-to-physician dialogues from metadata using LLMs and achieves high physician-rated quality in oncology and hepatology scenarios.
-
Anthropomorphism and Trust in Human-Large Language Model interactions
Warmth and cognitive empathy in LLMs drive higher anthropomorphism, trust, and relational closeness, especially on personal topics, while competence affects usefulness but not perceived human-likeness.
-
RECAP: Transparent Inference-Time Emotion Alignment for Medical Dialogue Systems
RECAP is an inference-time framework using cognitive appraisal theory to enhance emotional alignment and transparency in medical dialogue systems across model scales.
-
A Framework for Longitudinal Health AI Agents
Proposes a multi-layer framework and agent architecture that operationalizes adaptation, coherence, continuity, and agency for longitudinal health AI agents.
-
DR. INFO at the Point of Care: A Prospective Pilot Study of Physician-Perceived Value of an Agentic AI Clinical Assistant
A pilot study of 29 physicians found high perceived time savings (4.27/5) and decision support (4.16/5) from DR. INFO, yielding an NPS of 81.2 that dropped to 44.8 under conservative assumptions.