Current financial large language models(FinLLMs)exhibit two major limitations:the absence of standardized evaluation metrics for stock analysis quality and insufficient analytical depth.We address these limitations wi...Current financial large language models(FinLLMs)exhibit two major limitations:the absence of standardized evaluation metrics for stock analysis quality and insufficient analytical depth.We address these limitations with two contributions.First,we introduce AnalyScore,a systematic framework for evaluating the quality of stock analysis.Second,we construct Stocksis,an expert-curated dataset designed to enhance the financial analysis capabilities of large language models(LLMs).Building on Stocksis,together with a novel integration framework and quantitative tools,we develop FinSphere,an artificial intelligence(AI)agent that generates professional-grade stock analysis reports.Evaluations with AnalyScore show that FinSphere consistently surpasses general-purpose LLMs,domain-specific FinLLMs,and existing agent-based systems,even when the latter are enhanced with real-time data access and few-shot guidance.The findings highlight FinSphere’s significant advantages in analytical quality and real-world applicability.展开更多
Introduction:In-use product and material stocks are the amount of concerned manufactured products and materials in active use,and are essential components of urban ecosystem.Methods:This study estimates the dynamic in...Introduction:In-use product and material stocks are the amount of concerned manufactured products and materials in active use,and are essential components of urban ecosystem.Methods:This study estimates the dynamic in-use stocks of steel-containing products and steel in the city of Xiamen,China,during 1980–2015 by applying a bottom-up accounting approach.We incorporate 55 categories of steel-containing products that are classified into five end-use sectors(i.e.,buildings,infrastructure,transportation equipment,machinery,and domestic appliances).Outcomes and Discussion:In-use stocks of 51%of the studied products kept increasing during 1980–2015,especially after 2000.Steel stocks have grown up to 4.9±1.4 tons per capita(t/cap)in 2015,from 0.5±0.2 t/cap in 1980.Buildings are the largest reservoirs,although its share decreased from 89%in 1980 to 68%in 2015.The dynamic spatial distribution indicates that steel stocks gradually expanded from urban core to suburban areas.Conclusion:Theresults help to explore how a city’s urbanization is sustained by the in-use stocks growth.In-use steel stocks,of which the growth is highly correlated to and probably driven by the population growth,GDP increase,and urban built-up area expansion,may serve as a supplementary indicator for urbanization.展开更多
文摘Current financial large language models(FinLLMs)exhibit two major limitations:the absence of standardized evaluation metrics for stock analysis quality and insufficient analytical depth.We address these limitations with two contributions.First,we introduce AnalyScore,a systematic framework for evaluating the quality of stock analysis.Second,we construct Stocksis,an expert-curated dataset designed to enhance the financial analysis capabilities of large language models(LLMs).Building on Stocksis,together with a novel integration framework and quantitative tools,we develop FinSphere,an artificial intelligence(AI)agent that generates professional-grade stock analysis reports.Evaluations with AnalyScore show that FinSphere consistently surpasses general-purpose LLMs,domain-specific FinLLMs,and existing agent-based systems,even when the latter are enhanced with real-time data access and few-shot guidance.The findings highlight FinSphere’s significant advantages in analytical quality and real-world applicability.
基金This study was sponsored by the National Key Research and Development Program of Ministry of Science and Technology(2017YFC0505703)the Leading Project of Fujian Science and Technology Department(2016Y01010094)+2 种基金the Frontier Science Research Project of Chinese Academy of Sciences(QYZDB-SSW-DQC012)the Geological Surveying Project of China Geological Survey(No.121201103000150015)Wei-Qiang Chen acknowledges the Chinese Academy of Sciences Pioneer Hundred Talents Program.
文摘Introduction:In-use product and material stocks are the amount of concerned manufactured products and materials in active use,and are essential components of urban ecosystem.Methods:This study estimates the dynamic in-use stocks of steel-containing products and steel in the city of Xiamen,China,during 1980–2015 by applying a bottom-up accounting approach.We incorporate 55 categories of steel-containing products that are classified into five end-use sectors(i.e.,buildings,infrastructure,transportation equipment,machinery,and domestic appliances).Outcomes and Discussion:In-use stocks of 51%of the studied products kept increasing during 1980–2015,especially after 2000.Steel stocks have grown up to 4.9±1.4 tons per capita(t/cap)in 2015,from 0.5±0.2 t/cap in 1980.Buildings are the largest reservoirs,although its share decreased from 89%in 1980 to 68%in 2015.The dynamic spatial distribution indicates that steel stocks gradually expanded from urban core to suburban areas.Conclusion:Theresults help to explore how a city’s urbanization is sustained by the in-use stocks growth.In-use steel stocks,of which the growth is highly correlated to and probably driven by the population growth,GDP increase,and urban built-up area expansion,may serve as a supplementary indicator for urbanization.
