pith. sign in

arxiv: 2605.18770 · v1 · pith:DJH4DJMDnew · submitted 2026-04-15 · 💻 cs.IR · cs.AI

Agentic GraphRAG: Navigating Unstructured Financial Data with Collaborative AI

Arthur Capozzi , Dirk Helbing This is my paper

Pith reviewed 2026-05-21 00:40 UTC · model grok-4.3

classification 💻 cs.IR cs.AI
keywords GraphRAGagentic AIknowledge graphunstructured datacommercial registryinformation retrievalmulti-hop queriesmulti-turn conversation
0
0 comments X

The pith

An agentic GraphRAG system that merges structured registry records with LLM-extracted legal text outperforms standard vector-RAG on multi-hop and conversational queries.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper describes a framework for expert analysis of commercial registry data that combines structured fields with unstructured legal notices. It builds a Neo4j graph using deterministic ingestion for strong nodes, LLM extraction for weak nodes, and identity resolution before applying an analytical modular agent. The agent uses zero-shot intent routing, a bounded reflection loop, secure graph tools, and state-aware synthesis to handle complex questions. Evaluation on the Swiss Official Gazette of Commerce shows consistent gains over an agentic vector-RAG baseline in correctness, relevance, recall, turn success, and context carryover. A reader would care because public financial registries mix easy and hard data, and better navigation tools could speed up investigations while keeping human oversight.

Core claim

The central claim is that a three-phase pipeline of deterministic strong-node ingestion, LLM-based weak-node extraction from unstructured notices, and identity resolution produces a usable knowledge graph; when paired with a modular agent that performs zero-shot intent routing, bounded reflection, tool-mediated graph access, and state-aware response synthesis, the resulting agentic GraphRAG system delivers higher correctness, answer relevance, information recall, turn success rate, and context carryover accuracy than a standard agentic vector-RAG baseline across automated, human-curated, and conversational benchmarks on the Swiss Official Gazette of Commerce.

What carries the argument

The analytical modular agent that integrates zero-shot intent routing, a bounded reflection loop, secure tool-mediated graph access, and state-aware response synthesis, operating over a hybrid Neo4j knowledge graph built from structured fields and LLM-extracted weak nodes.

If this is right

  • The system supports more accurate multi-hop, temporal, and entity-centric investigations in public commercial registries than vector-only methods.
  • A human-in-the-loop dashboard exposes evidence and execution traces, enabling transparency and auditability for expert users.
  • The modular architecture transfers to other commercial gazettes and public-sector registry systems with similar mixed structured-unstructured data.
  • Performance advantages appear consistently across automated, human-curated, and multi-turn conversational evaluation tiers.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Hybrid graph construction that grounds LLM extractions in verified structured data may reduce hallucination risks in regulatory or legal retrieval tasks.
  • The framework could extend to other mixed-data domains such as medical claims or scientific patent records where entity resolution matters.
  • Replacing parts of the reflection loop with additional deterministic checks might further lower dependence on LLM quality without losing flexibility.

Load-bearing premise

The LLM extraction of entities and relations from unstructured legal notices must be accurate and complete enough that errors do not propagate through downstream agent queries and degrade final answers.

What would settle it

A manual review of the extracted weak nodes showing precision below the level needed for reliable multi-hop paths, followed by re-running the benchmarks where the GraphRAG system no longer outperforms the vector-RAG baseline on correctness or recall.

Figures

Figures reproduced from arXiv: 2605.18770 by Arthur Capozzi, Dirk Helbing.

Figure 1
Figure 1. Figure 1: Data ingestion pipeline for constructing the SHAB knowledge graph. Raw SHAB data combines unstructured [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Architecture of the analytical agent. The user query is first processed by a zero-shot intent router that restricts [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The Exploratory Dashboard operating in global search mode. The left navigation panel displays deterministic [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The Exploratory Dashboard operating in Dossier Mode. Since the user selected an entity in the left sidebar, [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The central dossier panel of the Exploratory Dashboard renders a 2D force-directed network to visualize [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
read the original abstract

We present a collaborative agentic GraphRAG framework for expert analysis of commercial registry data. Public registries are often formally accessible, yet difficult to use in practice because they combine structured records with large volumes of unstructured legal text. This limits conventional keyword and vector-only retrieval, especially for multi-hop, temporal, and entity-centric investigations. Our approach builds a Neo4j knowledge graph through a three-phase pipeline: (i) deterministic ingestion of strong nodes from verified structured fields, (ii) LLM-based extraction of weak nodes from unstructured notices, and (iii) deterministic identity resolution and deduplication. On top of this graph, we introduce an analytical modular agent that integrates zero-shot intent routing, a bounded reflection loop, secure tool-mediated graph access, and state-aware response synthesis. A human-in-the-loop dashboard exposes evidence and execution traces to support transparency and auditability. We evaluate the framework on the Swiss Official Gazette of Commerce, a multilingual corpus of more than seven million publications over seven years. We further contribute a multi-tier evaluation protocol covering entity-resolution precision, tool-routing behavior, answer quality, and multi-turn conversational performance. Across automated, human-curated, and conversational benchmarks, the proposed agentic GraphRAG system consistently outperforms a standard agentic vector-RAG baseline, with strong gains in correctness, answer relevance, information recall, turn success rate, and context carryover accuracy. The architecture is modular, reproducible, and transferable to other commercial gazettes and public-sector registry systems.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 3 minor

Summary. The manuscript presents Agentic GraphRAG, a collaborative agentic framework for expert analysis of commercial registry data combining structured and unstructured text. It constructs a Neo4j knowledge graph via a three-phase pipeline—deterministic ingestion of strong nodes from verified structured fields, LLM-based extraction of weak nodes from unstructured legal notices, and deterministic identity resolution and deduplication—then deploys an analytical modular agent with zero-shot intent routing, bounded reflection loop, secure tool-mediated graph access, and state-aware synthesis, supported by a human-in-the-loop dashboard. Evaluation on the Swiss Official Gazette of Commerce (over seven million multilingual publications) uses a multi-tier protocol covering entity-resolution precision, tool-routing, answer quality, and multi-turn performance; the agentic GraphRAG system is reported to consistently outperform a standard agentic vector-RAG baseline on correctness, answer relevance, information recall, turn success rate, and context carryover accuracy.

Significance. If the empirical claims hold, the work offers a modular, auditable architecture that could improve handling of multi-hop, temporal, and entity-centric queries over conventional vector RAG in regulatory and financial domains. The explicit separation of deterministic strong nodes from LLM-derived weak nodes, combined with the human-in-the-loop dashboard for traceability, addresses practical deployment concerns. The multi-tier evaluation protocol spanning automated, human-curated, and conversational benchmarks is a constructive contribution for assessing both retrieval and agentic components.

major comments (2)
  1. [three-phase pipeline description] Three-phase pipeline, phase (ii): the central claim that the hybrid graph supplies measurably better multi-hop and entity-centric evidence than pure vector retrieval rests on the quality of LLM-extracted weak nodes. No ablation isolating performance when the graph is built from strong nodes alone, nor any error-propagation analysis or controlled degradation of weak-node quality, is described. This leaves open whether downstream gains in correctness and recall survive when extraction errors occur.
  2. [evaluation protocol] Evaluation section / multi-tier protocol: the abstract asserts consistent outperformance with strong gains across metrics, yet supplies no quantitative results, confidence intervals, statistical tests, data-split details, or exclusion rules. Without these, the magnitude and reliability of improvements in answer relevance, turn success rate, and context carryover cannot be verified from the text.
minor comments (3)
  1. [analytical modular agent] Clarify how the bounded reflection loop interacts with the state-aware synthesis step to prevent infinite loops or context drift in multi-turn conversations.
  2. [entity-resolution precision metric] Add explicit comparison of entity-resolution precision against a non-LLM baseline to quantify the incremental value of the LLM extraction step.
  3. [conclusion / transferability paragraph] The claim of transferability to other commercial gazettes would benefit from a brief discussion of language-specific or jurisdiction-specific adaptations required for the deterministic ingestion and identity-resolution phases.

Simulated Author's Rebuttal

2 responses · 0 unresolved

Thank you for the detailed review and constructive feedback on our manuscript. We address each of the major comments below and outline the revisions we plan to make.

read point-by-point responses

  1. Referee: [three-phase pipeline description] Three-phase pipeline, phase (ii): the central claim that the hybrid graph supplies measurably better multi-hop and entity-centric evidence than pure vector retrieval rests on the quality of LLM-extracted weak nodes. No ablation isolating performance when the graph is built from strong nodes alone, nor any error-propagation analysis or controlled degradation of weak-node quality, is described. This leaves open whether downstream gains in correctness and recall survive when extraction errors occur.

    Authors: We acknowledge the importance of demonstrating the specific contribution of the LLM-extracted weak nodes. While our evaluation compares the full agentic GraphRAG system against an agentic vector-RAG baseline, which indirectly highlights the benefits of the graph structure including both strong and weak nodes, we agree that a dedicated ablation would provide stronger evidence. In the revised manuscript, we will include an ablation study comparing the full hybrid graph to a version built solely from strong nodes. Additionally, we will add an error-propagation analysis by introducing controlled noise into the weak node extraction and measuring impact on downstream metrics. revision: yes

  2. Referee: [evaluation protocol] Evaluation section / multi-tier protocol: the abstract asserts consistent outperformance with strong gains across metrics, yet supplies no quantitative results, confidence intervals, statistical tests, data-split details, or exclusion rules. Without these, the magnitude and reliability of improvements in answer relevance, turn success rate, and context carryover cannot be verified from the text.

    Authors: We thank the referee for pointing this out. The full paper contains detailed results from the multi-tier evaluation in Section 5, including specific performance numbers. However, to make the claims more verifiable directly from the abstract and to enhance transparency, we will revise the abstract to include key quantitative results with confidence intervals where applicable. We will also expand the evaluation section to explicitly include statistical tests, data-split details, and exclusion rules. This will allow readers to better assess the reliability of the reported improvements. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper presents an empirical framework consisting of a three-phase pipeline (deterministic strong-node ingestion, LLM weak-node extraction, deterministic deduplication) followed by an agentic query system evaluated on the external Swiss Official Gazette corpus against a distinct vector-RAG baseline. Central claims rest on reported gains in correctness, relevance, recall, and conversational metrics across automated, human-curated, and multi-turn benchmarks. No equations, self-definitional loops, fitted parameters renamed as predictions, or load-bearing self-citations appear in the derivation. The evaluation protocol and real-world corpus supply independent falsifiability outside any internal construction, rendering the results self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the reliability of LLM extraction from legal text and the correctness of deterministic identity resolution; no explicit free parameters or new physical entities are introduced in the abstract.

axioms (1)
  • domain assumption LLM-based extraction from unstructured legal notices yields sufficiently accurate entities and relations for graph construction
    Invoked in the second phase of the three-phase pipeline.

pith-pipeline@v0.9.0 · 5795 in / 1311 out tokens · 52288 ms · 2026-05-21T00:40:33.520336+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

52 extracted references · 52 canonical work pages · 1 internal anchor

  1. [1]

    Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG

    Aditi Singh, Abul Ehtesham, Saket Kumar, and Tala Talaei Khoei. Agentic retrieval-augmented generation: A survey on agentic rag.arXiv preprint arXiv:2501.09136, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  2. [2]

    From llm reasoning to autonomous ai agents: A comprehensive review, 2026

    Mohamed Amine Ferrag, Norbert Tihanyi, and Merouane Debbah. From llm reasoning to autonomous ai agents: A comprehensive review, 2026

    work page 2026

  3. [3]

    Structugraphrag: Structured document- informed knowledge graphs for retrieval-augmented generation.Proceedings of the AAAI Symposium Series, 4(1): 242–251, Nov

    Xishi Zhu, Xiaoming Guo, Shengting Cao, Shenglin Li, and Jiaqi Gong. Structugraphrag: Structured document- informed knowledge graphs for retrieval-augmented generation.Proceedings of the AAAI Symposium Series, 4(1): 242–251, Nov. 2024. doi: 10.1609/aaaiss.v4i1.31798

    work page doi:10.1609/aaaiss.v4i1.31798 2024

  4. [4]

    T-grag: A dynamic graphrag framework for resolving temporal conflicts and redundancy in knowledge retrieval

    Dong Li, Yichen Niu, Ying Ai, Xiang Zou, Biqing Qi, and Jianxing Liu. T-grag: A dynamic graphrag framework for resolving temporal conflicts and redundancy in knowledge retrieval. InProceedings of the 33rd ACM International Conference on Multimedia, MM ’25, page 11880–11889, New York, NY , USA, 2025. Association for Computing Machinery. ISBN 9798400720352....

    work page doi:10.1145/3746027.3755628 2025

  5. [5]

    Open data and transparency: Opportunities and challenges in the swiss context.Public Performance & Management Review, 43(3):662–686, 2020

    Tereza Cahlikova and Vincent Mabillard. Open data and transparency: Opportunities and challenges in the swiss context.Public Performance & Management Review, 43(3):662–686, 2020. doi: 10.1080/15309576.2019. 1657914

    work page doi:10.1080/15309576.2019 2020

  6. [6]

    Financial risk management and explainable, trustworthy, responsible ai.Frontiers in Artificial Intelligence, V olume 5 - 2022, 2022

    Sebastian Fritz-Morgenthal, Bernhard Hein, and Jochen Papenbrock. Financial risk management and explainable, trustworthy, responsible ai.Frontiers in Artificial Intelligence, V olume 5 - 2022, 2022. ISSN 2624-8212. doi: 10.3389/frai.2022.779799

    work page doi:10.3389/frai.2022.779799 2022

  7. [7]

    The role of explainable ai in enhancing trust and decision-making in financial services.Journal of Applied Finance & Banking, 15(5):49–62, 2025

    Ian Staley. The role of explainable ai in enhancing trust and decision-making in financial services.Journal of Applied Finance & Banking, 15(5):49–62, 2025

    work page 2025

  8. [8]

    Judging llm-as-a-judge with mt-bench and chatbot arena

    Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric Xing, Hao Zhang, Joseph E Gonzalez, and Ion Stoica. Judging llm-as-a-judge with mt-bench and chatbot arena. In A. Oh, T. Naumann, A. Globerson, K. Saenko, M. Hardt, and S. Levine, editors,Advances in Neural Information Processing Sys...

    work page 2023

  9. [9]

    2026 , issn =

    Jiawei Gu, Xuhui Jiang, Zhichao Shi, Hexiang Tan, Xuehao Zhai, Chengjin Xu, Wei Li, Yinghan Shen, Shengjie Ma, Honghao Liu, Saizhuo Wang, Kun Zhang, Zhouchi Lin, Bowen Zhang, Lionel Ni, Wen Gao, Yuanzhuo Wang, and Jian Guo. A survey on llm-as-a-judge.The Innovation, page 101253, 2026. ISSN 2666-6758. doi: https://doi.org/10.1016/j.xinn.2025.101253

    work page doi:10.1016/j.xinn.2025.101253 2026

  10. [10]

    Effective automatic feature engineering on financial statements for bankruptcy prediction

    Xinlin Wang, Zsófia Kräussl, Maciej Zurad, and Mats Brorsson. Effective automatic feature engineering on financial statements for bankruptcy prediction. In2023 3rd International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME), pages 1–8, 2023. doi: 10.1109/ICECCME57830. 2023.10252608

    work page doi:10.1109/iceccme57830 2023

  11. [11]

    Data mining techniques for the detection of fraudulent financial statements.Expert Systems with Applications, 32(4):995–1003, 2007

    Efstathios Kirkos, Charalambos Spathis, and Yannis Manolopoulos. Data mining techniques for the detection of fraudulent financial statements.Expert Systems with Applications, 32(4):995–1003, 2007. ISSN 0957-4174. doi: https://doi.org/10.1016/j.eswa.2006.02.016

    work page doi:10.1016/j.eswa.2006.02.016 2007

  12. [12]

    Bankruptcy prediction using case-based reasoning, neural networks, and discriminant analysis.Expert Systems with Applications, 13(2):97–108, 1997

    Hongkyu Jo, Ingoo Han, and Hoonyoung Lee. Bankruptcy prediction using case-based reasoning, neural networks, and discriminant analysis.Expert Systems with Applications, 13(2):97–108, 1997. ISSN 0957-4174. doi: https://doi.org/10.1016/S0957-4174(97)00011-0

    work page doi:10.1016/s0957-4174(97)00011-0 1997

  13. [13]

    Feature-weighted counterfactual-based explanation for bankruptcy prediction

    Soo Hyun Cho and Kyung shik Shin. Feature-weighted counterfactual-based explanation for bankruptcy prediction. Expert Systems with Applications, 216:119390, 2023. ISSN 0957-4174. doi: https://doi.org/10.1016/j.eswa.2022. 119390

    work page doi:10.1016/j.eswa.2022 2023

  14. [14]

    Jiaming Liu, Chengzhang Li, Peng Ouyang, Jiajia Liu, and Chong Wu. Interpreting the prediction results of the tree-based gradient boosting models for financial distress prediction with an explainable machine learning approach.Journal of Forecasting, 42(5):1112–1137, 2023. doi: https://doi.org/10.1002/for.2931

    work page doi:10.1002/for.2931 2023

  15. [15]

    Retrieval-augmented generation for knowledge-intensive nlp tasks

    Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. Retrieval-augmented generation for knowledge-intensive nlp tasks. In H. Larochelle, M. Ranzato, R. Hadsell, M.F. Balcan, and H. Lin, editors,Advances in Neural Informa...

    work page 2020

  16. [16]

    Evaluating retrieval-augmented generation models for financial report question and answering.Applied Sciences, 14(20), 2024

    Ivan Iaroshev, Ramalingam Pillai, Leandro Vaglietti, and Thomas Hanne. Evaluating retrieval-augmented generation models for financial report question and answering.Applied Sciences, 14(20), 2024. ISSN 2076-3417. doi: 10.3390/app14209318

    work page doi:10.3390/app14209318 2024

  17. [17]

    Finsage: A multi-aspect rag system for financial filings question answering

    Xinyu Wang, Jijun Chi, Zhenghan Tai, Tung Sum Thomas Kwok, Hailin He, Zhuhong Li, Yuchen Hua, Muzhi Li, Peng Lu, Suyucheng Wang, Yihong Wu, Huang Jerry, Jingrui Tian, Fengran Mo, Yufei Cui, and Ling Zhou. Finsage: A multi-aspect rag system for financial filings question answering. InProceedings of the 34th ACM International Conference on Information and K...

    work page doi:10.1145/3746252.3761587 2025

  18. [18]

    Hierfinrag—hierarchical multimodal rag for financial document understanding.Informatics, 13(2), 2026

    Quang-Vinh Dang, Ngoc-Son-An Nguyen, and Thi-Bich-Diem V o. Hierfinrag—hierarchical multimodal rag for financial document understanding.Informatics, 13(2), 2026. ISSN 2227-9709. doi: 10.3390/informatics13020030

    work page doi:10.3390/informatics13020030 2026

  19. [19]

    Large language models in finance: A survey,

    Boyu Zhang, Hongyang Yang, Tianyu Zhou, Muhammad Ali Babar, and Xiao-Yang Liu. Enhancing financial sentiment analysis via retrieval augmented large language models. InProceedings of the Fourth ACM International Conference on AI in Finance, ICAIF ’23, page 349–356, New York, NY , USA, 2023. Association for Computing Machinery. ISBN 9798400702402. doi: 10.1...

    work page doi:10.1145/3604237.3626866 2023

  20. [20]

    Mengshoel, and Xindong Wu

    Xiaoyi Fu, Xinqi Ren, Ole J. Mengshoel, and Xindong Wu. Stochastic optimization for market return prediction using financial knowledge graph. In2018 IEEE International Conference on Big Knowledge (ICBK), pages 25–32,

    work page

  21. [21]

    In2018 ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA)

    doi: 10.1109/ICBK.2018.00012. 26 Agentic GraphRAG

    work page doi:10.1109/icbk.2018.00012 2018

  22. [22]

    Knowledge graph-based event embedding framework for financial quantitative investments

    Dawei Cheng, Fangzhou Yang, Xiaoyang Wang, Ying Zhang, and Liqing Zhang. Knowledge graph-based event embedding framework for financial quantitative investments. InProceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’20, page 2221–2230, New York, NY , USA, 2020. Association for Computing Ma...

    work page doi:10.1145/3397271.3401427 2020

  23. [23]

    Anti-money laundering in bitcoin: Experimenting with graph convolutional networks for financial forensics.arXiv preprint arXiv:1908.02591, 2019

    Mark Weber, Giacomo Domeniconi, Jie Chen, Daniel Karl I Weidele, Claudio Bellei, Tom Robinson, and Charles E Leiserson. Anti-money laundering in bitcoin: Experimenting with graph convolutional networks for financial forensics.arXiv preprint arXiv:1908.02591, 2019

    work page arXiv 1908

  24. [24]

    Flowseries: flow analysis on financial networks.Applied Network Science, 10(1):28, 2025

    Arthur Capozzi, Salvatore Vilella, Dario Moncalvo, Marco Fornasiero, Valeria Ricci, Silvia Ronchiadin, and Giancarlo Ruffo. Flowseries: flow analysis on financial networks.Applied Network Science, 10(1):28, 2025. doi: 10.1007/s41109-025-00711-0

    work page doi:10.1007/s41109-025-00711-0 2025

  25. [25]

    Exploiting graph metrics to detect anomalies in cross-country money transfer temporal networks

    Salvatore Vilella, Arthur Thomas Edward Capozzi Lupi, Giancarlo Ruffo, Marco Fornasiero, Dario Moncalvo, Valeria Ricci, and Silvia Ronchiadin. Exploiting graph metrics to detect anomalies in cross-country money transfer temporal networks. InCompanion Proceedings of the ACM Web Conference 2023, WWW ’23 Companion, page 1245–1248, New York, NY , USA, 2023. A...

    work page doi:10.1145/3543873.3587602 2023

  26. [26]

    Weirdnodes: centrality based anomaly detection on temporal networks for the anti-financial crime domain.Applied Network Science, 10(1):14, 2025

    Salvatore Vilella, Arthur Capozzi, Marco Fornasiero, Dario Moncalvo, Valeria Ricci, Silvia Ronchiadin, and Giancarlo Ruffo. Weirdnodes: centrality based anomaly detection on temporal networks for the anti-financial crime domain.Applied Network Science, 10(1):14, 2025. doi: 10.1007/s41109-025-00702-1

    work page doi:10.1007/s41109-025-00702-1 2025

  27. [27]

    Temporal relational ranking for stock prediction.ACM Trans

    Fuli Feng, Xiangnan He, Xiang Wang, Cheng Luo, Yiqun Liu, and Tat-Seng Chua. Temporal relational ranking for stock prediction.ACM Trans. Inf. Syst., 37(2), March 2019. ISSN 1046-8188. doi: 10.1145/3309547

    work page doi:10.1145/3309547 2019

  28. [28]

    Beyond the tax haven: a graph analysis of business attraction in swiss municipalities.EPJ Data Science, 15, 2026

    Arthur Capozzi and Damian Dailisan. Beyond the tax haven: a graph analysis of business attraction in swiss municipalities.EPJ Data Science, 15, 2026. doi: 10.1140/epjds/s13688-026-00619-4

    work page doi:10.1140/epjds/s13688-026-00619-4 2026

  29. [29]

    Large language models in finance: A survey,

    Andy Chung and Kumiko Tanaka-Ishii. Modeling momentum spillover with economic links discovered from financial documents. InProceedings of the Fourth ACM International Conference on AI in Finance, ICAIF ’23, page 490–497, New York, NY , USA, 2023. Association for Computing Machinery. ISBN 9798400702402. doi: 10.1145/3604237.3626862

    work page doi:10.1145/3604237.3626862 2023

  30. [30]

    In-depth analysis of graph-based rag in a unified framework.Proc

    Yingli Zhou, Yaodong Su, Youran Sun, Shu Wang, Taotao Wang, Runyuan He, Yongwei Zhang, Sicong Liang, Xilin Liu, Yuchi Ma, and Yixiang Fang. In-depth analysis of graph-based rag in a unified framework.Proc. VLDB Endow., 18(13):5623–5637, January 2026. ISSN 2150-8097. doi: 10.14778/3773731.3773738

    work page doi:10.14778/3773731.3773738 2026

  31. [31]

    From local to global: A graph rag approach to query-focused summarization

    Darren Edge, Ha Trinh, Newman Cheng, Joshua Bradley, Alex Chao, Apurva Mody, Steven Truitt, Dasha Metropolitansky, Robert Osazuwa Ness, and Jonathan Larson. From local to global: A graph rag approach to query-focused summarization. 2025

    work page 2025

  32. [32]

    GRAG: Graph retrieval-augmented generation

    Yuntong Hu, Zhihan Lei, Zheng Zhang, Bo Pan, Chen Ling, and Liang Zhao. GRAG: Graph retrieval-augmented generation. In Luis Chiruzzo, Alan Ritter, and Lu Wang, editors,Findings of the Association for Computa- tional Linguistics: NAACL 2025, pages 4145–4157, Albuquerque, New Mexico, April 2025. Association for Computational Linguistics. ISBN 979-8-89176-19...

    work page doi:10.18653/v1/2025.findings-naacl.232 2025

  33. [33]

    Agentai: A comprehensive survey on autonomous agents in distributed ai for industry 4.0.Expert Systems with Applications, 291:128404, 2025

    Francesco Piccialli, Diletta Chiaro, Sundas Sarwar, Donato Cerciello, Pian Qi, and Valeria Mele. Agentai: A comprehensive survey on autonomous agents in distributed ai for industry 4.0.Expert Systems with Applications, 291:128404, 2025. ISSN 0957-4174. doi: https://doi.org/10.1016/j.eswa.2025.128404

    work page doi:10.1016/j.eswa.2025.128404 2025

  34. [34]

    A review of agentic artificial intelligence: Power of self-driven ai in the future of financial autonomy and enhanced customer engagement

    Aparna Krishna Bhat and Gokulram Krishnan. A review of agentic artificial intelligence: Power of self-driven ai in the future of financial autonomy and enhanced customer engagement. In2025 3rd International Conference on Sustainable Computing and Data Communication Systems (ICSCDS), pages 1160–1165, 2025. doi: 10.1109/ ICSCDS65426.2025.11167368

    work page arXiv 2025

  35. [35]

    Necessary but Not Perfect: Changes in AI Perception at a Large University

    Nir Kshetri. The rise of agentic ai in finance: Opportunities, risks, and human-centric integration.IT Professional, 27(4):19–24, 2025. doi: 10.1109/MITP.2025.3585227

    work page doi:10.1109/mitp.2025.3585227 2025

  36. [36]

    Elgendy, Mohamed Y .I

    Ibrahim A. Elgendy, Mohamed Y .I. Helal, Mohammed A. Al-Sharafi, Mousa Ahmed Albashrawi, Mohammad S. Al-Ahmadi, Il Jeon, and Yogesh K. Dwivedi. Agentic systems as catalysts for innovation in fintech: exploring opportunities, challenges and a research agenda.Information Discovery and Delivery, 05 2025. ISSN 2398-6247. doi: 10.1108/IDD-03-2025-0068

    work page doi:10.1108/idd-03-2025-0068 2025

  37. [37]

    Ai agents in finance and fintech: A scientific review of agent-based systems, applications, and future horizons.Computers, Materials and Continua, 86(1):1–34, 2025

    Maryan Rizinski and Dimitar Trajanov. Ai agents in finance and fintech: A scientific review of agent-based systems, applications, and future horizons.Computers, Materials and Continua, 86(1):1–34, 2025. ISSN 1546-2218. doi: https://doi.org/10.32604/cmc.2025.069678. 27 Agentic GraphRAG

    work page doi:10.32604/cmc.2025.069678 2025

  38. [38]

    Enhancing investment analysis: Optimizing ai-agent collaboration in financial research

    Xuewen Han, Neng Wang, Shangkun Che, Hongyang Yang, Kunpeng Zhang, and Sean Xin Xu. Enhancing investment analysis: Optimizing ai-agent collaboration in financial research. InProceedings of the 5th ACM International Conference on AI in Finance, ICAIF ’24, page 538–546, New York, NY , USA, 2024. Association for Computing Machinery. ISBN 9798400710810. doi: ...

    work page doi:10.1145/3677052.3698645 2024

  39. [39]

    Self-RAG: Learning to retrieve, gen- erate, and critique through self-reflection

    Akari Asai, Zeqiu Wu, Yizhong Wang, Avirup Sil, and Hannaneh Hajishirzi. Self-RAG: Learning to retrieve, gen- erate, and critique through self-reflection. InThe Twelfth International Conference on Learning Representations, 2024

    work page 2024

  40. [40]

    RAG-critic: Leveraging automated critic-guided agentic workflow for retrieval augmented generation

    Guanting Dong, Jiajie Jin, Xiaoxi Li, Yutao Zhu, Zhicheng Dou, and Ji-Rong Wen. RAG-critic: Leveraging automated critic-guided agentic workflow for retrieval augmented generation. In Wanxiang Che, Joyce Nabende, Ekaterina Shutova, and Mohammad Taher Pilehvar, editors,Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (...

    work page

  41. [41]

    ISBN 979-8-89176-251-0

    Association for Computational Linguistics. ISBN 979-8-89176-251-0. doi: 10.18653/v1/2025.acl-long.179

    work page doi:10.18653/v1/2025.acl-long.179 2025

  42. [42]

    Vincent Poor

    Ruichen Zhang, Hongyang Du, Yinqiu Liu, Dusit Niyato, Jiawen Kang, Sumei Sun, Xuemin Shen, and H. Vincent Poor. Interactive ai with retrieval-augmented generation for next generation networking.IEEE Network, 38(6): 414–424, 2024. doi: 10.1109/MNET.2024.3401159

    work page doi:10.1109/mnet.2024.3401159 2024

  43. [43]

    Hm-rag: Hierarchical multi-agent multimodal retrieval augmented generation

    Pei Liu, Xin Liu, Ruoyu Yao, Junming Liu, Siyuan Meng, Ding Wang, and Jun Ma. Hm-rag: Hierarchical multi-agent multimodal retrieval augmented generation. MM ’25, page 2781–2790, New York, NY , USA, 2025. Association for Computing Machinery. ISBN 9798400720352. doi: 10.1145/3746027.3754761

    work page doi:10.1145/3746027.3754761 2025

  44. [44]

    Ka-rag: Integrating knowledge graphs and agentic retrieval- augmented generation for an intelligent educational question-answering model.Applied Sciences, 15(23), 2025

    Fangqun Gao, Shu Xu, Weiyan Hao, and Tao Lu. Ka-rag: Integrating knowledge graphs and agentic retrieval- augmented generation for an intelligent educational question-answering model.Applied Sciences, 15(23), 2025. ISSN 2076-3417. doi: 10.3390/app152312547

    work page doi:10.3390/app152312547 2025

  45. [45]

    Keying hash functions for message authentication

    Mihir Bellare, Ran Canetti, and Hugo Krawczyk. Keying hash functions for message authentication. In Neal Koblitz, editor,Advances in Cryptology — CRYPTO ’96, pages 1–15, Berlin, Heidelberg, 1996. Springer Berlin Heidelberg. ISBN 978-3-540-68697-2

    work page 1996

  46. [46]

    Cypher: An evolving query language for property graphs

    Nadime Francis, Alastair Green, Paolo Guagliardo, Leonid Libkin, Tobias Lindaaker, Victor Marsault, Stefan Plantikow, Mats Rydberg, Petra Selmer, and Andrés Taylor. Cypher: An evolving query language for property graphs. InProceedings of the 2018 International Conference on Management of Data, SIGMOD ’18, page 1433–1445, New York, NY , USA, 2018. Associat...

    work page doi:10.1145/3183713.3190657 2018

  47. [47]

    React: Synergizing reasoning and acting in language models

    Shunyu Yao, Jeffrey Zhao, Dian Yu, Izhak Shafran, Karthik R Narasimhan, and Yuan Cao. React: Synergizing reasoning and acting in language models. InNeurIPS 2022 Foundation Models for Decision Making Workshop, 2022

    work page 2022

  48. [48]

    verbose database queries correlate with null results

    Kunlun Zhu, Zijia Liu, Bingxuan Li, Muxin Tian, Yingxuan Yang, Jiaxun Zhang, Pengrui Han, Qipeng Xie, Fuyang Cui, Weijia Zhang, et al. Where llm agents fail and how they can learn from failures.arXiv preprint arXiv:2509.25370, 2025

    work page arXiv 2025

  49. [49]

    Siva Reddy, Danqi Chen, and Christopher D. Manning. Coqa: A conversational question answering challenge. Transactions of the Association for Computational Linguistics, 7:249–266, 05 2019. ISSN 2307-387X. doi: 10.1162/tacl_a_00266

    work page doi:10.1162/tacl_a_00266 2019

  50. [50]

    URL https:// doi.org/10.18653/v1/d18-1241

    Eunsol Choi, He He, Mohit Iyyer, Mark Yatskar, Wen-tau Yih, Yejin Choi, Percy Liang, and Luke Zettlemoyer. QuAC: Question answering in context. In Ellen Riloff, David Chiang, Julia Hockenmaier, and Jun’ichi Tsujii, editors,Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2174–2184, Brussels, Belgium, October-No...

    work page doi:10.18653/v1/d18-1241 2018

  51. [51]

    Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers

    Wenhui Wang, Furu Wei, Li Dong, Hangbo Bao, Nan Yang, and Ming Zhou. Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers. InAdvances in Neural Information Processing Systems (NeurIPS), volume 33, pages 5776–5788, 2020

    work page 2020

  52. [52]

    RAGA s: Automated evaluation of retrieval augmented generation

    Shahul Es, Jithin James, Luis Espinosa Anke, and Steven Schockaert. RAGAs: Automated evaluation of retrieval augmented generation. In Nikolaos Aletras and Orphee De Clercq, editors,Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics: System Demonstrations, pages 150–158, St. Julians, Malta, March 202...

    work page doi:10.18653/v1/2024.eacl-demo.16 2024