Communication between Frozen Large Language Models via Prompt Optimization in a Referential Game
arXiv:2609.31989v1 Announce Type: new Abstract: We study communication between two frozen large language models from different providers, with different tokenizers, accessed through their API endpoints. The two play a referential game: one sees an object and describes it in a short fixed-length mes…
Read the full story at arXiv cs.CL ↗
Timeline · 2 reports
- 2026-09-29 04:00 · arXiv cs.CL
ScopeIF: Improving Scope-Aware Precise Instruction-Following in Large Language Models via Graded Reward Modeling - 2026-09-29 04:00 · arXiv cs.CL
Communication between Frozen Large Language Models via Prompt Optimization in a Referential Game