Search papers, labs, and topics across Lattice.
This study introduces BOUNDARY_SYNC, a novel protocol for quantifying representational coupling in multi-agent large language model (LLM) systems through the Coupling Amplification Factor (CAF). The researchers conducted controlled experiments using GPT-4o, revealing that text communication significantly homogenizes outputs (CAF=0.803), while image communication also leads to homogenization (CAF=0.834), though group size influences the direction of coupling. These findings highlight the measurable and controllable nature of LLM agent coupling, which has critical implications for the design of multi-agent systems.
Text communication among LLMs can lead to significant output homogenization, challenging assumptions about diversity in multi-agent interactions.
As large language models (LLMs) are deployed as communicating agents, does inter-agent communication cause outputs to converge? We introduce BOUNDARY_SYNC, a protocol measuring representational coupling via the Coupling Amplification Factor (CAF = JSD_cond / JSD_baseline), where CAF<1 indicates homogenization and CAF>1 indicates diversification. In controlled GPT-4o experiments (N=30, ~9,900 API calls), we measure coupling in text and image communication. Key findings: (1) text communication causes significant homogenization (CAF=0.803 [0.740, 0.873], d=1.30, p<0.001), confirmed by no-communication ablation and prompt-perturbation controls; (2) image communication also homogenizes under within-modality baselines (CAF=0.834 [0.811, 0.858]), with comparable proportional effect; (3) group size moderates coupling direction -- K=5 produces homogenization while K=3 yields CAF>1.0 (point estimates 1.14 and 1.06, CI pending), suggesting a directional shift toward diversification; (4) cross-model replication shows extreme variation (CAF 0.034-0.803), with DeepSeek dominated by format artifacts; (5) coupling is stateless -- driven by prompt context rather than cumulative updating, with continuous consensus producing monotonic convergence. These results establish LLM agent coupling as real, measurable, and controllable at the prompt level, with direct implications for multi-agent system design.
