Co-training both sides of a strategic dialogue with fully separated rewards makes each side generalize better to unseen counterparts than training against a fixed opponent.
Conversation policy.Use natural, concise, spoken language suitable for a phone call
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
IB-RL: Isolated Bilateral Reinforcement Learning for Strategic Dialogue Agents
Co-training both sides of a strategic dialogue with fully separated rewards makes each side generalize better to unseen counterparts than training against a fixed opponent.