There are some papers on artifical agents learning to communicate to play a game, etc. They either learn a language from scratch (an emergent language) or use reinforcement learning to improve an existing model of language.

• Multi-agent communication
  • 2020 - Multi-agent deep reinforcement learning: a survey

• Multi-agent
  • Havrylov & Titov 2017 - Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols
  • 2019 - Word-order biases in deep-agent emergent communication
  • 2020 - Emergent Multi-Agent Communication in the Deep Learning Era
  • 2020 - Open Problems in Cooperative AI
• Emergence of Grammar
  • Wechsler et al 2025 - The Emergence of Grammar through Reinforcement Learning

• Dialog
• Lifelong Learning
• Reinforcement Learning