Disentangling Topology and Diversity in Multi-Agent LLMs for Multilingual Low-Resource Emotion Detection

2026-09-13

AI Summary

研究通过独立分析推理拓扑和代理间多样性来源，解决多语言低资源情感检测问题，使用不同配置的多代理LLM系统进行评估。

Abstract

Multi-agent LLM systems combine multiple inference calls, but prior work often confounds how calls are connected with how they are diversified. We study these factors independently: inference topology and source of inter-agent diversity. In a controlled $2 \times 3$ matrix, we cross parallel aggregation and sequential refinement with stochastic sampling, role prompting, and learned QLoRA specialization, under a fixed three-call budget and output protocol within each backbone. Using Qwen2.5-14B-Instruct and Llama-3.1-8B-Instruct, we evaluate all six configurations on multilingual low-resource emotion detection across nine languages. Parallel learned specialization is strongest on Qwen at 52.83 Macro-F1 and reaches 52.94 on Llama. On Qwen it also exceeds same-backbone zero-shot, few-shot, CoT, and seven-call self-consistency baselines. The preferred topology depends on diversity source: sequential refinement helps stochastic and prompted settings, while the learned Width advantage shrinks from 2.83 points on Qwen to 0.17 on Llama. Depth-wise analysis suggests that later learned specialists can overwrite correct early predictions, although the aggregate effect is backbone-dependent. Overall, how agents are differentiated produces larger performance shifts than topology, which should be evaluated jointly with specialization.

Problem

• Multi-agent LLMs
• inference topology
• inter-agent diversity
• multilingual low-resource emotion detection

Innovation

• inference topology
• inter-agent diversity
• parallel learned specialization
• multilingual low-resource emotion detection
• QLoRA