用多智能体分工识别金融情绪，省钱又高效。
TriAgent: Divergence-Aware Multi-Agent Committees for Cost-Efficient Financial Sentiment Analysis
- 分词、句级、跨句三级智能体协同，按分歧度路由任务
- 在1.5B-7B模型上F1达0.87，比三人投票高21个百分点
- 适合追求低成本高精度金融分析的量化团队
生产级大模型金融情绪分析存在成本陷阱：多数查询可简单分类，但所有请求均由昂贵云推理引擎处理，费用随用户数线性增长。我们提出TriAgent，一种基于上下文粒度分层的多智能体委员会——词级词典（VADER）、句级领域模型（FinBERT）和跨句推理器（Qwen2.5，0.5B-14B-4bit，辅以Mistral-7B和Phi-3.5-mini跨家族验证）。三向语义分歧指数（SDI）衡量各粒度间分歧，据此路由每条查询。核心发现为‘批评者平台期’：当大模型被重用于评判小模型输出时，F1在1.5B-7B Qwen上稳定于~0.87（置信区间重叠），而同等规模三人投票仅达F1=0.66，源于粒度分层的多样性。由此衍生三项结论：(i) 基于多语言句向量的共享共识词典，使英文缓存可95%覆盖中文查询，F1=0.99——跨语言标准化零边际成本；(ii) SDI可作后验大模型幻觉检测，AUC=0.90；(iii) 单阶段SDI策略在20只股票回测中实现最优风险调整收益（夏普比3.50），优于始终使用FinBERT（1.36）和始终使用大模型（0.11）。在1000万用户规模下，相比GPT-4o-mini基准年省930万美元。代码、词典与共享共识词典已开源。
原文摘要 · Abstract (English)
Production LLM-based financial sentiment analysis faces a structural cost trap: most queries are trivially classifiable, yet expensive cloud reasoners process them all, and the bill scales linearly with user count. We present TriAgent, a multi-agent committee stratified by contextual granularity -- a word-level lexicon (VADER), a sentence-level domain transformer (FinBERT), and a cross-sentence reasoner (Qwen2.5, 0.5B-14B-4bit, with Mistral-7B and Phi-3.5-mini cross-family checks). A three-way Semantic Divergence Index (SDI) measures pairwise disagreement across granularities and routes each query accordingly. Our central finding is the critic plateau: when the LLM is re-tasked as a critic over the smaller agents' outputs, F1 plateaus at ~0.87 across 1.5B-7B Qwen (bootstrap 95% CIs overlap), while a same-size 3-persona vote drops to F1=0.66, which is driven by granularity-stratified diversity. Three corollaries follow from the same SDI signal: (i) a Shared Consensus Dictionary on multilingual sentence-BERT answers 95% of Chinese queries from an English cache at F1=0.99 -- cross-border canonicalization at zero marginal cost; (ii) SDI doubles as a post-hoc LLM-hallucination detector at AUC=0.90; (iii) the SDI single-stage strategy attains the best risk-adjusted return (Sharpe=3.50) on a 20-ticker back-test, dominating both always-FinBERT (1.36) and always-LLM (0.11). At 10M-user scale, TriAgent saves $9.3M/year vs. a GPT-4o-mini baseline. Code, lexicons, and the SCD are released.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。