A dual-agent co-training method using DPO on LLM-judged Pareto pairs and implicit adversarial reversal for the client simulator improves AI health coaching quality over one-sided baselines.
Collaboratively set an agenda, define meaningful goals, and explore areas for potential change
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Dual-Agent Co-Training for Health Coaching via Implicit Adversarial Preference Optimization
A dual-agent co-training method using DPO on LLM-judged Pareto pairs and implicit adversarial reversal for the client simulator improves AI health coaching quality over one-sided baselines.