Pith. sign in

REVIEW 3 major objections 4 minor 56 references

Explainable Information Retrieval in the Audit Domain

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Explainable information retrieval can assist auditors in generating complete audit reports, this position paper argues, and it maps the research directions and data obstacles needed to get there.

desk verdict A clean, honest position paper that maps XIR research directions for auditing, but the central benefit claim is explicitly admitted to be untested—worth a serious referee as a roadmap, not as a results paper. read the letter →

arxiv 2507.03479 v1 pith:2DEM6NGP submitted 2025-07-04 cs.IR

Alexander Frummet , Emanuel Slany , Jonas Amling , Moritz Lang , Stephan Scheele This is my paper
classification cs.IR
keywords explainableinformationretrievalauditneedsknowledgegraphsfinancialauditingconversationalagentsAIsearchtransparency
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper makes the case that explainable information retrieval (XIR) — search that shows how and why results were found — can assist auditors in generating complete audit reports. Auditors must gather evidence across general-ledger tables, financial reports, and approval records, and commercial conversational search agents that fabricate references are especially dangerous in this high-stakes setting. The authors argue that, as in cooking and web search, auditor information needs will range from simple fact-checking to complex investigative questions, and that the explanation a system gives should be matched to the type of need. They lay out three research directions — collecting auditor information needs, designing explanation formats, and representing audit knowledge as graphs — and three obstacles: closed expert environments, confidential data, and the lack of ground-truth explanations. If the agenda succeeds, annual audits of every major company could become more efficient, accurate, and effective at detecting fraud.

What carries the argument

The load-bearing conceptual machinery is the query-to-explanation matching principle adapted from domain-specific IR: the form an explanation should take is determined by the type of information need, so a system must first know whether an auditor is asking a factual lookup or an investigative chain question. The second mechanism is the knowledge graph as an evidence representation: because an audit item such as a revenue entry depends on underlying sales contracts, delivery records, and approval workflows, a graph can make the supporting chain visible, navigable, and explainable. The first mechanism decides how an explanation is presented, the second supplies what the explanation contains, and together they carry the paper's case that XIR can be purpose-built for auditing.

What would settle it

Collect real audit queries and explanation preferences from licensed auditors during an actual annual audit cycle, then test an explainable retrieval prototype that returns source-linked evidence chains; the central claim fails if auditor needs do not split into the predicted fact-checking and investigative types, or if the explainable system produces no measurable improvement in report completeness, accuracy, or expressed trust over a standard search interface.

Watch

Extended reading notes

Core claim

The paper's central claim is that explainable information retrieval — retrieval that returns not only documents but the reasons those documents are relevant — can support auditors in generating complete audit reports, and that auditing deserves dedicated XIR research rather than borrowing domain-agnostic methods. The authors reason from two observations: conversational search systems are becoming common but produce unreliable references, and auditing is a high-stakes, every-company task in which decisions based on bad information can be costly. Drawing on domain studies showing that simple and complex questions call for different answer and explanation styles, they predict auditor needs will split into fact-checking questions such as 'What was the invoice amount for transaction X?' and procedural or investigative questions such as 'How was this cost justified and approved across departments?'. They propose knowledge graphs as the natural representation for audit evidence because ledger entries, sales contracts, delivery records, and approval workflows form interdependent chains that a graph can display. The paper's contribution is this research agenda — the directions and the challenges — not a working prototype or an experiment.

Load-bearing premise

The argument rests on the analogy that auditors' information needs and explanation preferences will resemble those already mapped in cooking and web search, so the same query-to-explanation matching logic will transfer; the paper acknowledges in Section 3.1 that no realistic audit queries have been collected to test this.

Editorial extensions

If this is right

  • An XIR system that matches explanation format to query type could let an auditor verify a single invoice amount with a direct answer and, from the same interface, trace a cross-department approval chain through a visual evidence graph.
  • Representing audit documents as knowledge graphs would make the evidence behind any figure inspectable, for instance showing how a revenue entry is supported by contracts, delivery records, and approvals, which is the transparency high-stakes financial work requires.
  • Once auditor information needs are collected and categorized, retrieval systems could be designed to choose automatically between textual summaries, document links, and dependency visualizations for each task, reducing time spent searching general-ledger tables and financial reports.
  • Building and evaluating such systems will require ground-truth explanations for audit tasks, making multi-modal, post-hoc explanation methods tailored to different auditor roles a concrete research target.
  • If adopted, the approach could improve efficiency, accuracy, and fraud detection in annual audits, lowering costs and supporting financial integrity.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Our inference: the same evidence-chain logic likely transfers to neighboring high-stakes domains such as law and regulatory compliance, where conclusions also rest on linked documents and approval trails, even though the paper names only finance and audit.
  • Our inference: a fast empirical check of the core analogy would be to collect queries from practicing auditors during a real audit cycle and test whether they split into the predicted fact-checking and procedural-investigative categories; the paper itself notes that no realistic audit queries have been collected yet.
  • Our inference: the paper's sketch of anonymised expense-report data points to a tractable first experiment — a controlled comparison of textual summaries versus knowledge-graph visualizations on synthetic or anonymised audit data before any real confidential data becomes available.
  • Our inference: the framing carries an implicit normative claim the paper does not state — an LLM-based search tool whose references cannot be verified against a source chain should not be trusted as audit evidence at all.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper is a concise position statement arguing that explainable information retrieval (XIR), which is currently developed mostly for open-domain settings, should be extended to the audit domain. It motivates this with the problem of fabricated references in conversational agents and with the high-stakes nature of financial audits. Section 2 lists three research directions: understanding auditor information needs (2.1), designing explanations for audit tasks (2.2), and representing audit knowledge with knowledge graphs (2.3). Section 3 identifies three corresponding challenges: collecting realistic audit queries (3.1), accessing confidential real-world data (3.2), and defining ground-truth explanations for evaluation (3.3). The paper presents no experiments, datasets, or system prototypes; its central assertion, stated in Section 1, is that explainable IR systems can assist auditors in generating comprehensive audit reports.

Significance. The paper's value is as a research agenda rather than a validated contribution. If the central assertion were substantiated, the paper would open a new domain for XIR research and could have practical relevance for a regulated, high-stakes profession. The writing is clear, the references cover relevant XIR and XAI literature, and the authors are commendably explicit about the main gap: Section 3.1 admits that realistic audit queries have not been collected. The paper is honest about the difficulty of data access (Section 3.2) and about the context-dependence of explanations (Section 3.3). However, the benefit claim is currently a hypothesis based on an analogy with cooking and web search behavior, and the paper does not provide a concrete path to test it. This limits the current significance to agenda-setting.

major comments (3)
  1. [Section 2.1 and Section 3.1] The central claim, 'we believe that explainable IR systems can assist auditors in generating comprehensive audit reports' (Section 1), is load-bearing but rests on an untested analogy. Section 2.1 transfers Frummet and Elsweiler's cooking-domain distinction between fact-based and competence-oriented information needs to auditing ('it seems plausible'), while Section 3.1 explicitly states that realistic audit queries have not been collected because auditing is a closed, expert-driven environment. The examples 'invoice amount for transaction X' and 'how was this cost justified' are plausible but not evidence. I recommend reframing the assertion as a testable research hypothesis and describing concrete data-collection methods (e.g., interviews, diary studies, or partnerships with audit firms) that would validate the analogy.
  2. [Section 3.2] The practical benefit claim assumes that auditors or audit firms would adopt and act on explanations, but the paper does not address how explanations would fit professional audit standards or the evidentiary judgment required for an audit report. Section 3.2 notes that financial data are confidential and that anonymization may be needed, but it does not discuss whether synthetic data such as anonymized expense reports would preserve the multi-source evidence chains that Section 2.3 says are central. Transparent retrieval results are not the same as sufficient and appropriate audit evidence; the paper should either narrow its claim to information-retrieval assistance or engage with the evidentiary standards of auditing.
  3. [Section 3.3] The ground-truth explanation problem is identified but no evaluation methodology is proposed. Section 3.3 notes that explanation quality is context-dependent and that junior auditors and managers prefer different modalities, but it does not say how 'correct' explanations would be defined or measured in an audit setting. Without a way to evaluate whether an explanation improves audit outcomes (e.g., completeness of findings, time to identify inconsistencies), the research directions in Section 2 are not actionable. The authors should propose candidate evaluation metrics or study designs.
minor comments (4)
  1. [Section 2.1] The attribution of the fact-based vs. competence-oriented finding appears inconsistent: the sentence cites [15], whose bibliographic entry is 'Decoding the Metrics Maze...', while the finding on cooking-related information needs is described in [18] ('What Can I Cook with these Ingredients?...'). Please verify and correct the citation.
  2. [Reference [34]] The DOI for ExdocS is a placeholder ('https://doi.org/10.1145/nnnnnnn.nnnnnnn'); it should be completed or removed.
  3. [Section 3.1] The phrase 'crowdsourcing methods in a large scale' would read more naturally as 'crowdsourcing methods at scale'.
  4. [General] The paper would benefit from situating the proposal with respect to existing professional search and audit analytics tools (e.g., Verberne's explainable IR for professional search, or commercial audit software such as ACL and IDEA), clarifying what XIR adds beyond current exact-match lookups over ledgers and tables.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper is a position/roadmap piece with no derived predictions, and its self-citations are used only as cross-domain analogies rather than load-bearing evidence.

full rationale

This is a position paper that argues for future research directions rather than deriving results from fitted parameters or data. The central claim in Section 1, 'we believe that explainable IR systems can assist auditors in generating comprehensive audit reports,' is supported by general reasoning about audit complexity and the goals of XIR; it is not obtained by construction from any cited result. The cooking-domain findings of Frummet and Elsweiler [15] are invoked in Section 2.1 only as an analogy: 'Similarly, the audit domain likely involves a broad spectrum of information needs,' and the paper immediately qualifies the transfer as plausible rather than established ('It seems plausible that information needs in this domain could range from...'). Section 3.1 explicitly states that realistic audit queries have not been collected because auditing is a closed, expert-driven environment, which is an acknowledged limitation, not a circular justification. The self-citations [15-19] are empirical studies from the cooking domain and are used to motivate, not to force, the proposed research agenda; no equation is identified with another by construction, and no fitted input is renamed as a prediction. The paper is therefore self-contained as a research roadmap, and no significant circularity is present.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No free parameters or invented entities exist because the paper builds no models. The central argument rests on unvalidated domain assumptions about auditor work practices and data availability, which the authors acknowledge as challenges in Section 3.

assumptions (4)
  • domain assumption Every major company undergoes an annual audit, so a supporting XIR system could provide significant benefits.
    Invoked in Section 1 to motivate the application domain; no citation or evidence given for adoption scale.
  • domain assumption Auditors gather data from multiple sources such as SAP tables and financial reports and compile reports detailing findings.
    Section 1 describes the audit workflow; this grounds the proposed IR tasks but is not empirically sourced.
  • domain assumption Financial data is highly confidential and companies are unwilling or not allowed to share real-world data.
    Section 3.2 uses this to motivate anonymized datasets; it justifies a methodological constraint rather than a demonstrated fact.
  • domain assumption Knowledge graphs are well-suited for explainability because their structure conveys relationships between entities.
    Section 2.3 adopts this from cited prior work [6,9,25] and applies it to audit data without a pilot.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Explainable Information Retrieval in the Audit Domain." pith.science (2026). https://pith.science/paper/2DEM6NGP

@misc{pith2026250703479,
  author       = {Pith},
  title        = {Pith review of: Explainable Information Retrieval in the Audit Domain},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2DEM6NGP}},
  note         = {Machine review of arXiv:2507.03479}
}
read the original abstract

Conversational agents such as Microsoft Copilot and Google Gemini assist users with complex search tasks but often generate misleading or fabricated references. This undermines trust, particularly in high-stakes domains such as medicine and finance. Explainable information retrieval (XIR) aims to address this by making search results more transparent and interpretable. While most XIR research is domain-agnostic, this paper focuses on auditing -- a critical yet underexplored area. We argue that XIR systems can support auditors in completing their complex task. We outline key challenges and future research directions to advance XIR in this domain.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

56 extracted references · 29 canonical work pages

  1. [1]

    Abdalghani Abujabal, Rishiraj Saha Roy, Mohamed Yahya, and Gerhard Weikum

    work page

  2. [2]

    Avishek Anand, Lijun Lyu, Maximilian Idahl, Yumeng Wang, Jonas Wallat, and Zijian Zhang. 2022. Explainable Information Retrieval: A Survey. CoRR abs/2211.02405 (2022). https://doi.org/10.48550/arXiv.2211.02405

    arXiv doi:10.48550/arxiv.2211.02405 2022

  3. [3]

    Avishek Anand, Venktesh V, Abhijit Anand, and Vinay Setty. 2023. Query Understanding in the Age of Large Language Models. CoRR abs/2306.16004 (2023). https://doi.org/10.48550/arXiv.2306.16004

    arXiv doi:10.48550/arxiv.2306.16004 2023

  4. [4]

    Leif Azzopardi, Charles L. A. Clarke, Paul Kantor, Bhaskar Mitra, Johanne R. Trippas, Zhaochun Ren, Mohammad Aliannejadi, Negar Arabzadeh, Raman Chan- drasekar, Maarten de Rijke, Panagiotis Eustratiadis, William Hersh, Jin Huang, Evangelos Kanoulas, Jasmin Kareem, Yongkang Li, Simon Lupart, Kidist Amde Explainable Information Retrieval in the Audit Domain...

    work page arXiv 2024

  5. [5]

    Andrei Broder. 2002. A taxonomy of web search. SIGIR Forum 36, 2 (Sept. 2002), 3–10. https://doi.org/10.1145/792550.792552

    work page arXiv 2002

  6. [6]

    Boqi Chen, Kua Chen, Yujing Yang, Afshin Amini, Bharat Saxena, Cecilia Chávez- García, Majid Babaei, Amir Feizpour, and Dániel Varró. 2023. Towards Improving the Explainability of Text-based Information Retrieval with Knowledge Graphs. CoRR abs/2301.06974 (2023). https://doi.org/10.48550/arXiv.2301.06974

    arXiv doi:10.48550/arxiv.2301.06974 2023

  7. [8]

    Collins, Shervin Malmasi, Oleg Rokhlenko, and Eugene Agichtein

    Jason Ingyu Choi, Saar Kuzi, Nikhita Vedula, Jie Zhao, Giuseppe Castellucci, Marcus D. Collins, Shervin Malmasi, Oleg Rokhlenko, and Eugene Agichtein

    work page

  8. [9]

    Philipp Christmann, Rishiraj Saha Roy, and Gerhard Weikum. 2023. Explainable Conversational Question Answering over Heterogeneous Sources via Iterative Graph Neural Networks. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2023, Taipei, Taiwan, July 23-27, 2023 , Hsin-Hsi Chen, Wei-...

    arXiv 2023

Show all 56 references
  1. [10]

    Karen Church, Mauro Cherubini, and Nuria Oliver. 2014. A large-scale study of daily information needs captured in situ. ACM Trans. Comput.-Hum. Interact. 21, 2, Article 10 (Feb. 2014), 46 pages. https://doi.org/10.1145/2552193

    2014 doi

  2. [11]

    Shane Culpepper, Fernando Diaz, and Mark D

    J. Shane Culpepper, Fernando Diaz, and Mark D. Smucker. 2018. Research Frontiers in Information Retrieval: Report from the Third Strategic Workshop on Information Retrieval in Lorne (SWIRL 2018). SIGIR Forum 52, 1 (2018), 34–90. https://doi.org/10.1145/3274784.3274788

    2018

  3. [12]

    Eva Eigner and Thorsten Händler. 2024. Determinants of LLM-assisted Decision- Making. CoRR abs/2402.17385 (2024). https://doi.org/10.48550/arXiv.2402.17385

    2024 doi arXiv

  4. [13]

    Pichel, and David E

    Marcos Fernández-Pichel, Juan C. Pichel, and David E. Losada. 2025. Evaluating search engines and large language models for answering health questions. npj Digital Medicine 8, 1 (2025). https://doi.org/10.1038/s41746-025-01546-w

    2025 doi

  5. [14]

    Tafler, Stephan Scheele, and Ute Schmid

    Bettina Finzel, David E. Tafler, Stephan Scheele, and Ute Schmid. 2021. Expla- nation as a Process: User-Centric Construction of Multi-level and Multi-modal Explanations. In KI 2021: Advances in Artificial Intelligence , Stefan Edelkamp, Ralf Möller, and Elmar Rueckert (Eds.)....

    2021 doi

  6. [15]

    Alexander Frummet and David Elsweiler. 2024. Decoding the Metrics Maze: Nav- igating the Landscape of Conversational Question Answering System Evaluation in Procedural Tasks. In Proceedings of the Fourth Workshop on Human Evaluation of NLP Systems (HumEval) @ LREC-COLING 2024 ...

    2024

  7. [16]

    Alexander Frummet and David Elsweiler. 2024. QookA: A Cooking Question Answering Dataset. In Proceedings of the 2024 ACM SIGIR Conference on Human Information Interaction and Retrieval, CHIIR 2024, Sheffield, United Kingdom, March 10-14, 2024 , Paul D. Clough, Morgan Harvey, a...

    2024

  8. [17]

    Alexander Frummet, David Elsweiler, and Bernd Ludwig. 2019. Detecting Domain-specific Information needs in Conversational Search Dialogues. In Proceedings of the 3rd Workshop on Natural Language for Artificial Intelligence co-located with the 18th International Conference of t...

    2019

  9. [18]

    What Can I Cook with these Ingredients?

    Alexander Frummet, David Elsweiler, and Bernd Ludwig. 2022. "What Can I Cook with these Ingredients?" - Understanding Cooking-Related Information Needs in Conversational Search. ACM Trans. Inf. Syst. 40, 4 (2022), 81:1–81:32. doi:10.1145/3498330

    2022 doi

  10. [19]

    Alexander Frummet, Alessandro Speggiorin, David Elsweiler, Anton Leuski, and Jeff Dalton. 2024. Cooking with Conversation: Enhancing User Engagement and Learning with a Knowledge-Enhancing Assistant. ACM Trans. Inf. Syst. 42, 5, Article 122 (2024), 29 pages. https://doi.org/10...

    2024 doi

  11. [20]

    Shuyu Guo, Shuo Zhang, Weiwei Sun, Pengjie Ren, Zhumin Chen, and Zhaochun Ren. 2023. Towards Explainable Conversational Recommender Systems. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2023, Taipei, ...

    2023

  12. [21]

    Louisa Heidrich, Emanuel Slany, Stephan Scheele, and Ute Schmid. 2023. Fair- Caipi: A Combination of Explanatory Interactive and Fair Machine Learning for Human and Machine Bias Reduction. Machine Learning and Knowledge Extraction 5, 4 (2023), 1519–1538. doi:10.3390/make5040076

    2023 doi

  13. [22]

    Pattichis, and Douglas B

    Andreas Holzinger, Chris Biemann, Constantinos S. Pattichis, and Douglas B. Kell. 2017. What do we need to build explainable AI systems for the medical domain? CoRR abs/1712.09923 (2017). http://arxiv.org/abs/1712.09923

    2017 arXiv

  14. [23]

    Bernard J. Jansen. 2006. Search log analysis: What it is, what’s been done, how to do it. Library & Information Science Research 28, 3 (2006), 407–432. https://doi.org/10.1016/j.lisr.2006.06.005

    2006 doi

  15. [24]

    Jansen, Danielle Booth, and Brian Smith

    Bernard J. Jansen, Danielle Booth, and Brian Smith. 2009. Using the taxon- omy of cognitive learning to model online searching. Information Processing & Management 45, 6 (2009), 643–663. https://doi.org/10.1016/j.ipm.2009.05.004

    2009 doi

  16. [25]

    Zhen Jia, Soumajit Pramanik, Rishiraj Saha Roy, and Gerhard Weikum. 2021. Complex Temporal Question Answering on Knowledge Graphs. InCIKM ’21: The 30th ACM International Conference on Information and Knowledge Management, Virtual Event, Queensland, Australia, November 1 - 5, 2...

    2021

  17. [26]

    Diane Kelly, Jaime Arguello, Ashlee Edwards, and Wan-ching Wu. 2015. Devel- opment and Evaluation of Search Tasks for IIR Experiments using a Cognitive Complexity Framework. In Proceedings of the 2015 International Conference on The Theory of Information Retrieval (Northampton...

    2015

  18. [27]

    Hyun-Woo Kim, Dong-Hyeon Shin, Jiyoung Kim, Gha-Hyun Lee, and Jae Wook Cho. 2024. Assessing the performance of ChatGPT’s responses to questions related to epilepsy: A cross-sectional study on natural language processing and medical information retrieval. Seizure - European Jou...

    2024 doi

  19. [28]

    Weronika Lajewska, Damiano Spina, Johanne Trippas, and Krisztian Balog. 2024. Explainability for Transparent Conversational Information-Seeking. In Proceed- ings of the 47th International ACM SIGIR Conference on Research and Develop- ment in Information Retrieval, SIGIR 2024, ...

    2024

  20. [29]

    Jurek Leonhardt, Koustav Rudra, and Avishek Anand. 2023. Extractive Explana- tions for Interpretable Text Ranking.ACM Trans. Inf. Syst. 41, 4 (2023), 88:1–88:31. https://doi.org/10.1145/3576924

    2023 doi

  21. [30]

    Michael Llordes, Debasis Ganguly, Sumit Bhatia, and Chirag Agarwal. 2023. Explain Like I am BM25: Interpreting a Dense Model’s Ranked-List with a Sparse Approximation. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retr...

    2023

  22. [31]

    Lundberg and Su-In Lee

    Scott M. Lundberg and Su-In Lee. 2017. A Unified Approach to Interpreting Model Predictions. In Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, December 4-9, 2017, Long Beach, CA, USA , Isabelle Guyon, Ulri...

    2017

  23. [32]

    Tri Nguyen, Mir Rosenberg, Xia Song, Jianfeng Gao, Saurabh Tiwary, Rangan Majumder, and Li Deng. 2016. MS MARCO: A Human Generated MAchine Reading COmprehension Dataset. In Proceedings of the Workshop on Cogni- tive Computation: Integrating neural and symbolic approaches 2016 ...

    2016

  24. [33]

    Kosuke Nishida, Kyosuke Nishida, Masaaki Nagata, Atsushi Otsuka, Itsumi Saito, Hisako Asano, and Junji Tomita. 2019. Answering while Summarizing: Multi-task Learning for Multi-hop QA with Evidence Extraction. In Proceedings of the 57th Conference of the Association for Computa...

    2019

  25. [34]

    Sayantan Polley, Atin Janki, Marcus Thiel, Juliane Hoebel-Mueller, and Andreas Nuernberger. 2021. ExdocS: Evidence Based Explainable Document Search. InThe 1st International Workshop on Causality in Search and Recommendation (CSR’21) Frummet et al. (Online). CEUR Workshop Proc...

    2021 doi

  26. [35]

    Why Should I Trust You?

    Marco Túlio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016. "Why Should I Trust You?": Explaining the Predictions of Any Classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, August 13-17, 2...

    2016

  27. [36]

    Avi Rosenfeld. 2021. Better Metrics for Evaluating Explainable Artificial In- telligence. In AAMAS ’21: 20th International Conference on Autonomous Agents and Multiagent Systems, Virtual Event, United Kingdom, May 3-7, 2021 , Frank Dignum, Alessio Lomuscio, Ulle Endriss, and A...

    2021

  28. [37]

    Rydzewski, Deepak Dinakaran, Shuang G

    Nicholas R. Rydzewski, Deepak Dinakaran, Shuang G. Zhao, Eytan Ruppin, Baris Turkbey, Deborah E. Citrin, and Krishnan R. Patel. 2024. Comparative Evaluation of LLMs in Clinical Oncology. NEJM AI 1, 5 (2024). https://doi.org/10.1056/ aioa2300151

    2024

  29. [38]

    Gesina Schwalbe and Bettina Finzel. 2023. A comprehensive taxonomy for explainable artificial intelligence: a systematic survey of surveys on methods and concepts. Data Mining and Knowledge Discovery (2023). doi:10.1007/s10618- 022-00867-8

    2023 doi

  30. [39]

    Jaspreet Singh and Avishek Anand. 2019. EXS: Explainable Search Using Local Model Agnostic Interpretability. In Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining, WSDM 2019, Melbourne, VIC, Aus- tralia, February 11-15, 2019 , J. Shane Culpep...

    2019 doi

  31. [40]

    Emanuel Slany, Stephan Scheele, and Ute Schmid. 2024. Bayesian CAIPI: A Probabilistic Approach to Explanatory and Interactive Machine Learning. InArti- ficial Intelligence. ECAI 2023 International Workshops , Sławomir Nowaczyk, Prze- mysław Biecek, Neo Christopher Chung, Mauro...

    2024

  32. [41]

    White, Reid Andersen, Georg Buscher, Sathish Manivannan, Nagu Rangan, and Longqi Yang

    Siddharth Suri, Scott Counts, Leijie Wang, Chacha Chen, Mengting Wan, Tara Safavi, Jennifer Neville, Chirag Shah, Ryen W. White, Reid Andersen, Georg Buscher, Sathish Manivannan, Nagu Rangan, and Longqi Yang. 2024. The Use of Generative Search Engines for Knowledge Work and Co...

    2024 doi arXiv

  33. [42]

    Stefano Teso and Kristian Kersting. 2019. Explanatory Interactive Machine Learning. In Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society, AIES 2019, Honolulu, HI, USA, January 27-28, 2019 , Vincent Conitzer, Gillian K. Hadfield, and Shannon Vallor (Eds.). ...

    2019

  34. [43]

    Suzan Verberne. 2018. Explainable IR for Personalizing Professional Search. In Joint Proceedings of the First International Workshop on Professional Search (ProfS2018); the Second Workshop on Knowledge Graphs and Semantics for Text Retrieval, Analysis, and Understanding (KG4IR...

    2018

  35. [44]

    Mittelstadt, and Chris Russell

    Sandra Wachter, Brent D. Mittelstadt, and Chris Russell. 2017. Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPR. (2017). arXiv:1711.00399

    2017 arXiv

  36. [45]

    Walters and Esther Isabelle Wilder

    William H. Walters and Esther Isabelle Wilder. 2023. Fabrication and errors in the bibliographic citations generated by ChatGPT. Scientific Reports 13, 1 (2023), 14045. https://doi.org/10.1038/s41598-023-41032-5

    2023 doi

  37. [46]

    Yumeng Wang, Xiuying Chen, and Suzan Verberne. 2024. QUIDS: Query Intent Generation via Dual Space Modeling. CoRR abs/2410.12400 (2024). https: //doi.org/10.48550/arXiv.2410.12400

    2024 doi

  38. [47]

    d’Avila Garcez

    Adam White and Artur S. d’Avila Garcez. 2020. Measurable Counterfactual Local Explanations for Any Classifier. In ECAI 2020 - 24th European Conference on Artificial Intelligence, 29 August-8 September 2020, Santiago de Compostela, Spain, August 29 - September 8, 2020 - Includi...

    2020 doi

  39. [48]

    Ryen W. White. 2023. Tasks, Copilots, and the Future of Search: A Keynote at SIGIR 2023. SIGIR Forum 57, 2 (2023), 4:1–4:8. doi:10.1145/3642979.3642985

    2023

  40. [49]

    Barbara Wildemuth, Luanne Freund, and Elaine G. Toms. 2014. Untangling search task complexity and difficulty in the context of interactive information retrieval studies. Journal of Documentation 70, 6 (2014), 1118–1140. https: //doi.org/10.1108/JD-03-2014-0056

    2014 doi

  41. [50]

    Haoyi Xiong, Jiang Bian, Yuchen Li, Xuhong Li, Mengnan Du, Shuaiqiang Wang, Dawei Yin, and Sumi Helal. 2024. When Search Engine Services Meet Large Lan- guage Models: Visions and Challenges. IEEE Transactions on Services Computing 17, 6 (2024), 4558–4577. doi:10.1109/TSC.2024.3451185

    2024

  42. [51]

    Zhilin Yang, Peng Qi, Saizheng Zhang, Yoshua Bengio, William Cohen, Ruslan Salakhutdinov, and Christopher D. Manning. 2018. HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language...

    2018

  43. [52]

    Puxuan Yu, Razieh Rahimi, and James Allan. 2022. Towards Explainable Search Results: A Listwise Explanation Generator. In SIGIR ’22: The 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, Madrid, Spain, July 11 - 15, 2022 , Enrique Am...

    2022

  44. [53]

    ChengXiang Zhai. 2024. Large Language Models and Future of Information Retrieval: Opportunities and Challenges. In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR ’24). Association for Computing Machinery, ...

    2024

  45. [54]

    Ruqing Zhang, Jiafeng Guo, Yixing Fan, Yanyan Lan, and Xueqi Cheng. 2020. Query Understanding via Intent Description Generation. In Proceedings of the 29th ACM International Conference on Information & Knowledge Management (Virtual Event, Ireland) (CIKM ’20). Association for C...

    2020

  46. [55]

    Huaqin Zhao, Zhengliang Liu, Zihao Wu, Yiwei Li, Tianze Yang, Peng Shu, Shaochen Xu, Haixing Dai, Lin Zhao, Hanqi Jiang, Yi Pan, Junhao Chen, Yifan Zhou, Gengchen Mai, Ninghao Liu, and Tianming Liu. 2024. Revolutionizing Finance with LLMs: An Overview of Applications and Insig...

    2024

  47. [2017]

    QUINT: Interpretable Question Answering over Knowledge Bases. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, EMNLP 2017, Copenhagen, Denmark, September 9-11, 2017 - System Demonstrations, Lucia Specia, Matt Post, and Michael Paul (Ed...

    2017 doi

  48. [2022]

    Wizard of Tasks: A Novel Conversational Dataset for Solving Real-World Tasks in Conversational Settings. In Proceedings of the 29th International Confer- ence on Computational Linguistics, COLING 2022, Gyeongju, Republic of Korea, October 12-17, 2022, Nicoletta Calzolari, Chu-...

    2022

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.