# Selective and Hybrid-State Communication for Latent Multi-Agent Systems

Preprint · Miles Liao, Aaron Smieszek, Yuhao Ge

- Archive id: arxiv:2610.37779
- Published: 2026-10-27
- arXiv: https://arxiv.org/abs/2610.37779
- Source URL: https://arxiv.org/abs/2610.37779
- PDF: https://arxiv.org/pdf/2610.37779
- Provenance: arxiv

## Abstract

Efficient communication remains a major bottleneck in large language model multiagent systems. Latent communication reduces repeated natural-language exchange by transferring internal model states between agents, but these transfers remain costly and are typically designed around conventional KV-cache representations. We address this problem from two complementary perspectives. We introduce SelectKV, a boundary-aware method that selectively communicates KV states while protecting inherited context, and HyLaMAS, which extends latent collaboration to hybrid linear-attention models by communicating both KV caches and recurrent states through a shared workspace. Across six benchmarks, SelectKV reduces communicated KV positions and logical KV payload by 9.7-9.9%, with accuracy unchanged on two benchmarks, within one percentage point on three, and six points lower on HumanEval+. Across Qwen3.5 model scales, HyLaMAS reduces decoded output tokens by 53-80% and achieves 1.16-2.78x speedups relative to text-based multi-agent communication. Together, these independently evaluated methods represent two complementary approaches to efficient latent-state communication across conventional Transformer and hybrid model architectures.

_Abstract reproduced under CC0-1.0 (arXiv descriptive metadata)._

Topics: [Efficient Training & Inference](https://space-kitty.com/topics/efficient-ai)

---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.