Interpretable Emergent Language Using Inter-Agent Transformers

Mannan Bhardwaj
5/4/2025
cs.AIcs.CL

No official code found. Best implementation: MannanB/DIAT (1 star, last commit 2025-05-06).

Abstract

This paper explores the emergence of language in multi-agent reinforcement learning (MARL) using transformers. Existing methods such as RIAL, DIAL, and CommNet enable agent communication but lack interpretability. We propose Differentiable Inter-Agent Transformers (DIAT), which leverage self-attention to learn symbolic, human-understandable communication protocols. Through experiments, DIAT demonstrates the ability to encode observations into interpretable vocabularies and meaningful embeddings, effectively solving cooperative tasks. These results highlight the potential of DIAT for interpretable communication in complex multi-agent environments.

View on arXivView PDF

Code Implementations(1)

MannanB/DIATCommunity100%
10PythonMar 14, 20251 years ago

Similar Papers

Towards Language-Augmented Multi-Agent Deep Reinforcement Learning

87% match2025

Scalable Communication for Multi-Agent Reinforcement Learning via Transformer-Based Email Mechanism

85% match2023

Verco: Learning Coordinated Verbal Communication for Multi-agent Reinforcement Learning

84% match2024

MACTAS: Self-Attention-Based Module for Inter-Agent Communication in Multi-Agent Reinforcement Learning

83% match2025

Language Grounded Multi-agent Reinforcement Learning with Human-interpretable Communication

83% match1 repo182024

Towards Efficient Collaboration via Graph Modeling in Reinforcement Learning

83% match2024

Cite this paper

@article{bhardwaj2025interpretable,
  title  = {Interpretable Emergent Language Using Inter-Agent Transformers},
  author = {Mannan Bhardwaj},
  year   = {2025},
  eprint = {2505.02215},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2505.02215v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2505.02215.svg)](https://codeofpaper.com/paper/2505.02215)

Discussion