Manalyzer: End-to-end Automated Meta-analysis with Multi-agent System
Paper • 2505.20310 • Published • 1
field stringclasses 1
value | paper_idx stringclasses 111
values | doi stringclasses 111
values | type stringclasses 3
values | table_or_image imagewidth (px) 333 2.14k ⌀ | text_or_caption stringlengths 1 115k |
|---|---|---|---|---|---|
atmosphere | 0000 | 10.1016/j.envpol.2018.04.071 | table | Table 1 Summary of the mean concentrations of $\mathrm{PM}_{2.5}$ , OC, and EC $\left(\upmu\mathrm{g}\:\mathsf{m}^{-3}\right)$ in Shenzhen during the controlled and uncontrolled periods at two sampling sites, along with the concentrations of SOA for isoprene, $\mathfrak{x}_{}$ -pinene, $\upbeta.$ -caryophyllene and tol... | |
atmosphere | 0000 | 10.1016/j.envpol.2018.04.071 | image | Fig. 1. Wind rose plots (a) showing the frequency of wind directions in Shenzhen during the controlled period and uncontrolled periods, along with time series of the daily ambient concentrations of $\mathrm{PM}_{2.5}$ at Longgang and Peking University and visibility (b); time series of carbonaceous aerosol (OC, EC, and... | |
atmosphere | 0000 | 10.1016/j.envpol.2018.04.071 | image | Fig. 2. Source apportionment of $\mathrm{PM}_{2.5}\ \mathrm{OC}$ in Shenzhen at Longgang (LG) and Peking University (PU), reported as the average relative source contributions to OC $(\%)$ during controlled and uncontrolled periods. | |
atmosphere | 0000 | 10.1016/j.envpol.2018.04.071 | image | Fig. 3. Ambient concentrations of secondary organic tracers: A) sum of three isoprene SOA tracers, B) the sum of four $\mathfrak{x}$ -pinene SOA tracers, and C) one toluene SOA tracer at LG and PU during the controlled and uncontrolled periods. | |
atmosphere | 0000 | 10.1016/j.envpol.2018.04.071 | image | Fig. 4. Comparison of the fossil and contemporary sources of total carbon (equivalent to the sum of organic and elemental carbon) estimated by a) radiocarbon $(^{14}\mathrm{C})$ analysis, and b) chemical mass balance (CMB) modeling. | |
atmosphere | 0000 | 10.1016/j.envpol.2018.04.071 | text | Not supported with pagination yet | Source apportionment of fine particulate matter organic carbon in Shenzhen, China by chemical mass balance and radiocarbon methods\*
Ibrahim M. Al-Naiema a, Subin Yoon b, Yu-Qin Wang c, d, Yuan-Xun Zhang c, e, Rebecca J. Sheesley b, \*, Elizabeth A. Stone a, f, \*\*
a Department of Chemistry, University of Iowa, Iowa C... |
atmosphere | 0001 | 10.1080/10962247.2012.701193 | image | Figure 1. $\mathrm{PM}_{2.5}$ samples were taken in seven southern China cities: Chongqing (CQ), Guangzhou (GZ), Hong Kong (HK), Hangzhou (HZ), Shanghai (SH), Wuhan (WH), and Xiamen (XM); and seven northern China cities: Beijing (BJ), Changchun (CC), Jinchang (JC), Qingdao (QD), Tianjin (TJ), Xi’an (XA), and Yulin (YL)... | |
atmosphere | 0001 | 10.1080/10962247.2012.701193 | image | Figure 2. Average (square), median (central horizontal bar), 25th and 75th percentiles (lower and upper bars), 1st and 99th percentiles (lower and upper x), and minimum and maximum $(-)$ concentrations for each chemical component across all cities and seasons. Average chemical components are ordered by abundance, with ... | |
atmosphere | 0001 | 10.1080/10962247.2012.701193 | table | Table 1. Arithmetic averages standard deviations (mg m3) for PM2.5 mass and chemical components by city and season. See Figure 1 for city codes. Each average contains 14 values | |
atmosphere | 0001 | 10.1080/10962247.2012.701193 | table | Table 2. Comparison of $\mathrm{PM}_{2.5}$ chemical component ratios for the 14 Chinese cities with ratios from selected cities in Europe, Canada, Mexico, and the United States | |
atmosphere | 0001 | 10.1080/10962247.2012.701193 | image | Figure 3. Relationships between $\mathrm{PM}_{2.5}$ As, $\mathrm{Pb}$ , and $\mathrm{SO}_{4}{}^{2-}$ concentrations from the 14 cities during winter and summer, 2003. | |
atmosphere | 0001 | 10.1080/10962247.2012.701193 | image | Figure 4. Wintertime material balance of $\mathrm{PM}_{2.5}$ for the 14 Chinese cities. Organic matter (OM) is estimated as $1.6\times\mathrm{OC}$ (Chen and Yu, 2007; El-Zanan et al., 2005; ElZanan et al., 2009) to account for unmeasured hydrogen and oxygen. Geological material is estimated as $25\times\mathrm{Fe}$ (Ca... | |
atmosphere | 0001 | 10.1080/10962247.2012.701193 | image | Figure 5. Summertime material balance of $\mathrm{PM}_{2.5}$ for the 14 Chinese cities. Organic matter, geological material, and others are explained in the Figure 4 caption. | |
atmosphere | 0001 | 10.1080/10962247.2012.701193 | table | Table 3. Comparison of PM2.5 and major chemical concentrations (mg m3) from this study with measurements from other PM2.5 studies in Beijing (BJ), Xi’an (XA), Shanghai (SH), and Guangzhou (GZ) | |
atmosphere | 0001 | 10.1080/10962247.2012.701193 | text | Not supported with pagination yet | Journal of the Air & Waste Management Association
Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uawm20
Winter and Summer $\mathbf{PM}_{2.5}$ Chemical Compositions in Fourteen Chinese Cities
Jun-Ji Cao a , Zhen-Xing Shen b , Judith C. Chow a c , John... |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | image | Fig. 1. Location of the sampling site at Xi’an, China. | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | table | Table 1. Average of OC and EC concentrations during September 2003 to February 2004 at Xi’an, China. | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | image | Fig. 2. Time series of $\mathrm{PM}_{2.5}$ mass, organic carbon (OC), elemental carbon (EC), fraction of $\mathrm{PM}_{2.5}$ composed of $\mathrm{OC}\!\times\!1.6\!+\!\mathrm{EC}$ $(\mathrm{TCA}\%)$ , and OC/EC ratios at Xi’an from 13 September 2003 to 29 February 2004. OC is multiplied by 1.6 for the $\mathrm{TCA}\%$ ... | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | image | Fig. 3. Relationships between OC and EC concentrations in $\mathrm{PM}_{2.5}$ and $\mathrm{PM_{10}}$ . | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | image | Fig. 4. Distribution of $\mathrm{PM}_{2.5}$ and $\mathrm{PM_{10}}$ mass concentrations during fall and winter. The valid paired samples were 17 in fall and 36 in winter. The box plots indicate the mean $24\mathrm{-h}$ concentration and the min, 1st, 25th, 50th, 75th, 99th and max percentiles. A normal curve is fitted t... | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | table | Table 2. Statistical summary of the percentage of OC, EC, and $\mathrm{TCA}\%$ in $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{10}^{\mathrm{a}}$ | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | image | Fig. 5. Abundances (mass fraction of total carbon) of eight thermally-derived carbon fractions in ambient and source samples. | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | table | Table 3. Comparison of $\operatorname{PM}_{2.5}$ OC, EC at Xi’an with other Asian cities. | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | image | Fig. 6. Periodicity of $\mathrm{PM}_{2.5}$ OC, EC, mass, and daily average wind speed. (PSD TISA on the $\mathrm{Y}$ axis refers to Power as Time-Integral Squared Amplitude.) | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | table | Table 4. APCA results of fall samples. | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | table | Table 5. APCA results of winter samples. | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | image | Fig. 7. Relative contributions of major sources to $\mathrm{PM}_{2.5}$ TC during fall and winter 2003. | |
atmosphere | 0002 | 10.5194/acp-5-3127-2005 | text | Not supported with pagination yet | Characterization and source apportionment of atmospheric organic and elemental carbon during fall and winter of 2003 in Xi’an, China
J. J. $\mathbf{Cao}^{1}$ , F. $\mathbf{W}\mathbf{u}^{1,2}$ , J. C. Chow3, S. C. Lee4, Y. Li1, S. W. Chen5, Z. S. $\mathbf{A}\mathbf{n}^{1}$ , K. K. Fung6, J. G. Watson3, C. S. $\mathbf{Zh... |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | image | Fig. 1. Sampling site and its surroundings in a rural area in Lingcheng $37^{\circ}21^{\prime}17^{\prime\prime}\mathrm{N}$ , $116^{\circ}28^{\prime}30^{\prime\prime}\mathrm{E}$ ), a district of Dezhou City in Shandong Province, China. | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | image | Fig. 2. CMAQ modeling domains at a horizontal grid resolution of $27\,\mathrm{km}$ over China (D1, with 180 columns and 150 rows, $\sim\!1.97\times10^{7}\,\mathrm{km}^{2})$ and $9\,\mathrm{km}$ over an area in northern China (D2, with 120 columns and 111 rows, $\sim\!1.08\times10^{6}\,\mathrm{km}^{2};$ . The zoom-in ar... | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | table | Table 1 Descriptive statistics of chemical species in $\mathrm{PM}_{2.5}$ in terms of concentrations $(\upmu\mathrm{g}/\uppi^{3})$ and percentages (in brackets, $\%$ ). | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | table | Table 2 Average concentrations of $\mathrm{PM}_{2.5}$ , $S0_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ in Lingcheng and other areas in China. | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | table | Table 3 The mass concentration of secondary organic carbon (SOC) during the sampling period. | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | image | Fig. 3. Temporal variations in OC and EC abundances $\left(\upmu\mathrm{g}/\uppi^{3}\right)$ and OC/EC ratios at the sampling site in Lingcheng. | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | table | Table 4 Performance statistics for $\mathrm{PM}_{2.5}$ , OC, EC, $S0_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ concentrations. | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | image | Fig. 4. Scatter plots of the daily simulated versus observed concentrations of $\mathrm{PM}_{2.5},$ OC, EC, $S0_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ during the winter sampling period in 2010. The daily simulated concentrations were calculated by the averaging the hourly simulated results from the... | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | image | Fig. 5. Comparison between daily simulated and observed $\mathrm{PM}_{2.5}$ , $S0_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , $\mathrm{NH}_{4}^{+}$ , OC and EC at the Lingcheng study site from November 21st to December 20th. Observations are shown with solid line, and simulations are shown with dashed line. The daily simulated... | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | table | Table 5 Average contributions of $\mathrm{PM}_{2.5}$ and main species from local (Lingcheng) and surrounding regions during winter and heavy haze days (in brackets) $(\%)$ . | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | image | Fig. 6. Percent contributions of $\mathrm{PM}_{2.5}$ from the four directions (north, east, west, and south; the simulation area is equally divided into four parts centered on the sampling site). | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | image | Fig. 7. The contribution of $\mathrm{PM}_{2.5}$ per unit area (contribution $/\mathrm{km}^{2}$ ). | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | image | Fig. 8. 12-Hour backward trajectories reaching the sampling site for each hour on 21–24 November and 7, 8, 16, 17, and 21 December on a Lambert conformal projection map of North China. | |
atmosphere | 0003 | 10.1016/j.scitotenv.2017.01.066 | image | Fig. 9. Comparison of the $\mathrm{PM}_{2.5}$ contribution rates during the period of relatively clean days $(\mathrm{PM}_{2.5}\leq75\;\upmu\mathrm{g}/\mathrm{m}^{3})$ , haze days $(75~|\mathrm{{ug/m^{3}}<\mathrm{{PM_{2.5}}<200~|\mathrm{{ug/m^{3}}})}}$ and heavy haze days $(\mathrm{PM}_{2.5}\geq200\,\upmu\mathrm{g}/\ma... |
End of preview. Expand in Data Studio
Manalyzer: End-to-end Automated Meta-analysis with Multi-agent System
🆕 Updates/News
🚩 Updates (2026-02-09) Code has been released.
🚩 Updates (2025-05-22) Initial upload to arXiv [PDF]. The code will be released soon.
🎯 Abstract
Meta-analysis is a systematic research methodology that synthesizes data from multiple existing studies to derive comprehensive conclusions. This approach not only mitigates limitations inherent in individual studies but also facilitates novel discoveries through integrated data analysis. Traditional meta-analysis involves a complex multi-stage pipeline including literature retrieval, paper screening, and data extraction, which demands substantial human effort and time. However, while LLM-based methods can accelerate certain stages, they still face significant challenges, such as hallucinations in paper screening and data extraction. In this paper, we propose a multi-agent system, Manalyzer, which achieves end-to-end automated meta-analysis through tool calls. The hybrid review, hierarchical extraction, self-proving, and feedback checking strategies implemented in Manalyzer significantly alleviate these two hallucinations. To comprehensively evaluate the performance of meta-analysis, we construct a new benchmark comprising 729 papers across 3 domains, encompassing text, image, and table modalities, with over 10,000 data points. Extensive experiments demonstrate that Manalyzer achieves significant performance improvements over the LLM baseline in multi meta-analysis tasks.
🚀 Method Overview
Manalyzer is a multi-agent system incorporating tool calling and feedback mechanisms, enabling end-to-end automated meta-analysis in real scientific research scenarios. We divide the meta-analysis process into three stages. The first stage involves receiving user input, searching for and downloading papers, followed by filtering out relevant and valuable ones. The second stage focuses on extracting data from these selected papers and integrating it into tables. The third stage is to analyze the integrated data and output the final meta-analysis report.
🔥 Quick Start
export LLM_API_KEY="your-api-key"
export LLM_BASE_URL="your-api-base-url"
export MINERU_TOKEN="your-mineru-api-key" # Apply for the API at https://mineru.net/
python workflow/main.py
📬 Contact Us
💬 GitHub Issues: Please open an issue for bug reports or feature requests
📧 Email: xu_wanghan@sjtu.edu.cn
📜 Citation
If you would like to cite our work, please use the following BibTeX.
@article{xu2025manalyzer,
title={Manalyzer: End-to-end Automated Meta-analysis with Multi-agent System},
author={Xu, Wanghan and Zhang, Wenlong and Ling, Fenghua and Fei, Ben and Hu, Yusong and Ren, Fangxuan and Lin, Jintai and Ouyang, Wanli and Bai, Lei},
journal={arXiv preprint arXiv:2505.20310},
year={2025}
}
🌟 Star History
If you find this work helpful, please consider to star⭐ this repo. Thanks for your support! 🤩
- Downloads last month
- 65
Size of downloaded dataset files:
779 MB
Size of the auto-converted Parquet files:
779 MB
Number of rows:
6,657