LLMs trained with self-play negotiation achieve collective agency alignment comparable to single-agent baselines while substantially improving conflict-resolution performance.
Question the initial premises
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Learning to Negotiate: Multi-Agent Deliberation for Collective Value Alignment in LLMs
LLMs trained with self-play negotiation achieve collective agency alignment comparable to single-agent baselines while substantially improving conflict-resolution performance.