Pith. sign in

REVIEW 4 major objections 5 minor 50 references

GoT-CD: Graph-of-Thoughts Causal Discovery and the Fragility of Post-hoc Path-Specific Fairness Audits

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read A discovered graph that misses the protected-attribute pathway produces a confident zero in a path-specific fairness audit, indistinguishable from genuine fairness.

desk verdict The fairness-fragility warning is correct and the GoT-CD design is a genuine variation, but the reported numbers are internally inconsistent and need repair before the empirical claim is credible. read the letter →

arxiv 2608.02877 v1 pith:5N5M5CZE submitted 2026-08-03 cs.LG

Nitish Nagesh , Elahe Khatibi , Thomas Dean Hughes , Mahdi Bagheri , Pratik Gajane , Amir M. Rahmani This is my paper · ORCID
classification cs.LG
keywords causaldiscoverylargelanguagemodelsgraph-of-thoughtspath-specificfairnesscounterfactualDAGvalidityAlzheimer'sdiseaseclinicaldecisionsupport
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 claims that a causal graph discovered from data can look structurally accurate yet be badly wrong for the fairness question it is meant to answer. It introduces GoT-CD, a language-model-based discovery method whose reasoning unit is a complete candidate edge set, and shows that on five benchmarks GoT-CD returns valid DAGs with the best structural F1 among LLM methods on three of them. On an Alzheimer's benchmark with a known unfair pathway, however, five of eight discovered graphs—including structurally competitive ones—contain no path from the protected attribute to the outcome and therefore report a path-specific effect of exactly zero, indistinguishable from a genuinely fair result. The paper argues that chains of discovery-plus-audit can certify unfair models as fair, and that evaluation should score survival of the specific pathway the audit depends on, not aggregate structural agreement alone.

What carries the argument

The load-bearing object is the candidate edge set as the unit of reasoning in a Graph-of-Thoughts pass: Generate produces k complete graphs in parallel, Score applies a deterministic validity function, KeepBestN retains the best candidates, Aggregate merges them under a hard union constraint that forbids edges no reasoning branch proposed, Improve refines the merged graph, and a final greedy projection guarantees a DAG. The fairness side rests on the path-specific effect: for each simple directed path π from protected attribute S to outcome Y, $PSE(\pi)=\prod_{l=0}^{L-1} B_{v_l,v_{l+1}}$, and $PSE_{|\cdot|}=\sum_\pi |PSE(\pi)|$ measures total path mass. The mechanism that produces the false-clean result is that a graph with no $S\to Y$ path yields exactly zero path-specific effect by construction, the same number a genuinely fair model would produce.

What would settle it

Run the locked Alzheimer's protocol ($n=100$, the same backbone, $k=3$, seed 0) ten times and record $F_1$ and path-specific effect per run; if any graph with no $S\to Y$ path returns a nonzero or flagged estimate, the confident-zero mechanism fails, and if the two reported $F_1$ values for the same configuration (0.757 in Table 1 and 0.811 in Table 3) do not reproduce, the runs are not stable enough to support the ranking.

Watch

Extended reading notes

Core claim

The central claim is that path-specific fairness audits are only as good as the graph they are handed, and aggregate structural scores do not reveal when the relevant sub-structure is missing. On the Alzheimer's benchmark, GoT-CD recovers the ground-truth unfair path Sex→Brain Volume→MOCA Score, giving a total linear effect of −0.580 against the benchmark's −0.572, while LLM-BFS, with a competitive F1 of 0.649, recovers no S→Y path at all and reports PSE equal to zero and zero discrete effects. Because this zero is produced by structural omission, it is a false-clean certificate rather than evidence of fairness. GES shows the complementary failure: it finds the path but inflates total path mass to 4.034 via eight spurious routes, versus a true value of 0.572. The paper concludes that discovery-to-fairness pipelines should report pathway recovery alongside the fairness estimate, and that discovery methods intended for such pipelines should be evaluated on the edges the downstream analysis depends on.

Load-bearing premise

The load-bearing premise is that a single locked run with one language-model backbone, one temperature, and one random seed is representative of how the methods compare, and that the hand-written validity scorer and the order-dependent greedy DAG projection do not dominate the reported rankings.

Editorial extensions

If this is right

  • For any pipeline that chains causal discovery to path-specific fairness, reporting whether the $S\to Y$ pathway survived discovery is a minimal safeguard; without it, an omitted pathway is indistinguishable from a genuinely fair result.
  • Ranking discovery methods by aggregate structural F1 can mislead fairness users: LLM-BFS scores 0.649 on Alzheimer's yet misses the only unfair path, while GoT-CD at 0.757 recovers it.
  • Always-acyclic output matters for auditability: GoT-CD returns valid DAGs on all five benchmarks, whereas PC is cyclic on all five and GES on four, forcing heuristic edge-breaking before path-specific estimands are well-defined.
  • Discovery methods fail in opposite directions: a missing pathway produces a confident zero, while spurious recovered pathways can inflate unfairness mass, as GES does with PSE equal to 4.034 against a true 0.572.
  • The gain of GoT-CD comes from full-graph reasoning rather than Graph-of-Thoughts machinery as such, since the GoT-within-BFS hybrid underperforms and misses the unfair pathway on Alzheimer's.

Reading between the lines

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

  • The confident-zero mechanism is not specific to fairness: any estimand defined over a designated path, such as a mediation effect or a policy-relevant causal pathway, will inherit a missing pathway as an exact zero rather than a flagged failure; the paper's Alzheimer's case is one instance of a general property.
  • A natural testable extension is to replace the deterministic validity scorer with a data-based score such as BIC or conditional-independence statistics while keeping the global thought structure; the paper's own Sweden-Traffic results suggest this would improve pathway recovery where the language-model prior alone fails.
  • The order-dependent greedy DAG projection could be made more stable by ordering candidate edges by scorer reliability or confidence rather than list position, reducing the arbitrariness of which edge in a proposed cycle survives.
  • Running the locked protocol across multiple seeds, temperatures, and backbones would test whether the divergence between structural and fairness fidelity is a property of these methods or of this particular protocol; the paper leaves that question open.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes GoT-CD, an LLM-based causal discovery method that treats complete candidate edge sets as 'thoughts' in a Graph-of-Thoughts pipeline, scores them with a deterministic validity function, merges candidates under a hard union constraint, and enforces acyclicity by greedy projection. It reports structural benchmarks on five datasets at n=100 with gpt-4o-mini, claiming DAG-valid outputs everywhere and the best DAG-valid F1 among LLM baselines on Asia, Alzheimer's, and COVID-Respiratory. On an Alzheimer's benchmark with the designated unfair path Sex→Brain Volume→MOCA Score, the paper applies a post-hoc path-specific fairness audit and finds that five of eight discovered graphs contain no S→Y path and hence report PSE=0, while GoT-CD recovers the path and GES inflates path mass; it concludes that structural fidelity does not imply fairness fidelity and recommends reporting pathway recovery alongside fairness estimates.

Significance. The 'confident zero' failure mode is a real and practically important point: since path-specific effects are sums over directed paths in the supplied graph, a missing protected-attribute pathway produces an estimate of exactly zero rather than a flagged uncertainty. This observation is partly definitional, and the paper states it correctly in Section 6.3; the value added lies in the empirical demonstration that real discovery methods actually produce such graphs and in the proposed GoT-CD method. The method design—full-graph reasoning units, deterministic scoring, and a hard union constraint—is a reasonable contribution, and the paper is transparent about several limitations in Section 6.4 (single backbone, single sample size, simulated data). However, the central empirical claims currently rest on internally inconsistent, single-run results with no code or data, so the paper's main demonstration cannot yet be verified.

major comments (4)
  1. [Tables 2-3 and Section 6.1] The paper's locked protocol (k=3, seed 0, n=100) gives different results in different tables for every dataset: Table 2 reports GoT-CD F1 = 0.750 on Asia, 0.757 on Alzheimer's, 0.688 on COVID-R, and 0.111 on Sweden-T, while Table 3's k=3 row reports 0.632, 0.811, 0.727, and 0.286, respectively; Table 5 reports two S→Y paths with PSE|·|=1.031 for GoT-CD on Alzheimer's, whereas Section 6.1 describes the k=3 setting as recovering exactly one path with PSE|·|=0.582. Because all runs are claimed to be locked, these discrepancies are not a matter of error bars; the reader cannot determine which graph is the actual output of the method. Since the paper's central empirical claim is that GoT-CD recovers the fairness-relevant pathway while five baselines do not, this contradiction undermines the main contribution and must be resolved with a consistent protocol, code, data, and preferably multiple seeds.
  2. [Section 4] All reported results are single runs with no standard errors, no multiple seeds, and no code or data release. LLM generation at temperature 0.7 is stochastic, and a fixed client-side seed does not by itself guarantee reproducible API outputs across versions, so the observed path-recovery result could be a single lucky draw. The paper should report repeated runs (or at least error bars and full candidate logs) before claiming that GoT-CD stably recovers the unfair path or that five of eight methods reliably produce confident zeros.
  3. [Section 3.2] The acyclicity guarantee is achieved by a final greedy projection whose output is order-dependent: the paper itself states that the surviving edge in any proposed cycle is determined by position in the emitted list rather than by evidential support. No tie-breaking rule, random order, or sensitivity analysis is provided. Because the projected DAG is exactly the graph used for the fairness audit, this arbitrary ordering could determine whether the S→Y pathway survives, so the central comparison requires either a deterministic tie-breaking rule with justification or results aggregated over many orderings.
  4. [Section 3.2] The deterministic validity scorer has several hand-chosen parameters (undeclared-variable penalty -5, self-loop penalty -3, acyclicity bonus +3, cyclic penalty -4, and density bounds 0.5n to 2.5n), and no sensitivity analysis is reported except for the branch factor k. With only single runs, these arbitrary weights could dominate the behavior of the method, so the paper does not yet establish that full-graph reasoning topology, rather than the specific scorer and projection choices, is responsible for the reported performance.
minor comments (5)
  1. [References [1] and [2]] Reference [2] is a verbatim duplicate of reference [1] but is cited as the source of the 'recent collection of applied causal graphs' and the Alzheimer's reference graph; this citation cannot be correct and must be fixed.
  2. [Section 4.3] The k=5 row on Sweden-Traffic is reported as undefined F1 (p+r=0), meaning GoT-CD outputs an empty graph at k=5; this deserves an explanation and a discussion of what it implies for the method's behavior.
  3. [Section 3.3] The GoT-CD-BFS hybrid is described only briefly; please specify how candidate variable sets are generated, scored, and merged so the ablation is reproducible.
  4. [Abstract and Sections 1.5/6.1] The wording that full-graph reasoning 'yields always-acyclic discovered graphs' overstates the mechanism, since acyclicity is imposed by post-processing rather than by the reasoning process itself; rephrase for accuracy.
  5. [General] The paper would benefit from a data and code availability statement; currently no link or repository is mentioned.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the fragility result is a direct consequence of the PSE definition and is presented as such, while the empirical demonstration is independent of any fitted parameters.

full rationale

The paper's central claim is that post-hoc path-specific fairness audits inherit discovery errors in a way that is not flagged: a graph missing the S→Y pathway yields PSE=0. This is not a circular prediction but a direct logical consequence of the estimand's definition (Eq. 2 defines PSE as a sum over directed paths in the supplied graph; an empty path set gives zero). The paper explicitly labels this 'by construction' (Section 6.1: 'yields PSE |·|=0 and NDE=NIE=0 by construction') and uses it as a warning about pipeline fragility, not as an empirical discovery derived from the method's outputs. GoT-CD's structural results and pathway recovery are produced by LLM generation, a deterministic hand-specified scorer, and greedy projection; no parameter is fitted to the fairness estimand, and the Alzheimer's benchmark graph is an external reference. The only self-citation (ALCM, ref. [15], by two of the present authors) appears in related work as background and is not load-bearing; no uniqueness theorem or ansatz is imported from the authors' prior work. The acknowledged order-dependence of the greedy projection and the absence of error bars are reproducibility/correctness concerns, not circularity. Therefore no circular step is exhibited.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central claims rest on a handful of hand-chosen scorer constants, a reconstructed reference graph for Sweden-Traffic, and parametric assumptions (linear-Gaussian SEM, no unobserved confounding) for the fairness audit. No new entities are introduced.

free parameters (5)
  • Validity scorer edge penalty for undeclared variables = -5
    Hand-chosen constant in the deterministic scorer (Section 3.2); affects candidate ranking.
  • Validity scorer self-loop penalty = -3
    Hand-chosen in Section 3.2.
  • Acyclicity bonus / cyclic penalty = +3 / -4
    Hand-chosen in Section 3.2.
  • Density prior bounds = 0.5n and 2.5n edges
    Soft prior in Section 3.2; n likely number of variables; thresholds hand-chosen.
  • Branch factor k = 3 (ablated 1, 5)
    Main setting chosen by hand; ablation does not show a clear optimum across datasets.
assumptions (5)
  • domain assumption The reference graphs (Asia, Child, Alzheimer's, COVID-Respiratory, Sweden-Traffic) are correct ground-truth causal structures.
    Structural F1 and path-recovery scores are computed against these references; Sweden-Traffic was reconstructed from a published edge table (Section 3.1).
  • domain assumption Alzheimer's data are generated from a linear-Gaussian SEM parameterized by the reference graph, so OLS estimation of path coefficients is correctly specified under the discovered graph.
    Section 3.5 and Section 6.4 acknowledge this; real clinical data would not guarantee this.
  • domain assumption The path-specific fairness framework of Wu et al. applies with the supplied DAG, and no unobserved confounding beyond the graph.
    Section 3.5; standard assumption in causal fairness.
  • domain assumption The LLM's pretrained knowledge is a valid source of causal structure at n=100.
    Premise of LLM-based discovery in the cited works [13,16].
  • ad hoc to paper Greedy projection edge order does not materially affect the conclusions.
    Section 3.2 acknowledges order-dependence; Section 6.4 lists it as a limitation. The paper proceeds without quantifying its effect on F1 or path recovery.

how reviews work

0 comments
Cite this review

Pith. "Pith review of GoT-CD: Graph-of-Thoughts Causal Discovery and the Fragility of Post-hoc Path-Specific Fairness Audits." pith.science (2026). https://pith.science/paper/5N5M5CZE

@misc{pith2026260802877,
  author       = {Pith},
  title        = {Pith review of: GoT-CD: Graph-of-Thoughts Causal Discovery and the Fragility of Post-hoc Path-Specific Fairness Audits},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5N5M5CZE}},
  note         = {Machine review of arXiv:2608.02877}
}
read the original abstract

Causal discovery recovers directed structure from observational data and is increasingly used in clinical settings to support mechanism reasoning and fairness audits of predictive models. Path-specific counterfactual fairness asks whether a protected attribute influences an outcome through illegitimate pathways, but these estimands are defined relative to a supplied causal graph and therefore inherit whatever errors the discovery step introduces. Discovery methods are routinely scored on aggregate structural metrics that weight all edges equally, and no established evaluation asks whether the specific pathway an audit depends on survives discovery---or what the audit reports when that pathway is missing. Here we show that full-graph Graph-of-Thoughts reasoning yields acyclic discovered graphs that are structurally competitive with large language model (LLM) baselines, yet that structural fidelity alone does not guarantee fairness-faithful audits. We introduce GoT-CD, in which the reasoning unit is a complete candidate edge set: multiple graphs are generated in parallel, scored by a deterministic validity function, and merged under a hard union constraint that forbids invented edges, with greedy projection enforcing a DAG before commitment. GoT-CD returns a valid DAG on all five reported benchmarks and achieves the best DAG-valid F1 score among LLM methods on Asia, Alzheimer's, and COVID-Respiratory datasets. On an Alzheimer's benchmark with known unfair path, a post-hoc path-specific audit shows that five of eight discovered graphs recover no path from the sensitive attribute to the outcome and therefore report a null overall effect while mediated effects persist, necessitating downstream path-specific fairness analysis along with structural discovery.

Figures

Figures reproduced from arXiv: 2608.02877 by the authors.

Figure 1
Figure 1. Proposed evaluation pipeline (reported suite). Five benchmarks—Asia, Child, COVID [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 39 canonical work pages

  1. [1]

    Causal modelling agents: causal graph discov- ery through synergising metadata- and data-driven reasoning

    Ahmed Abdulaal, Nina Montana-Brown, Tiantian He, Ayodeji Ijishakin, Ivana Drobnjak, Daniel C Castro, Daniel C Alexander, et al. Causal modelling agents: causal graph discov- ery through synergising metadata- and data-driven reasoning. InThe Twelfth International Conference on Learning Representations. ICLR, 2023

    work page 2023

  2. [2]

    Causal modelling agents: causal graph discovery 15 through synergising metadata- and data-driven reasoning

    Ahmed Abdulaal, Nina Montana-Brown, Tiantian He, Ayodeji Ijishakin, Ivana Drobnjak, Daniel C Castro, Daniel C Alexander, et al. Causal modelling agents: causal graph discovery 15 through synergising metadata- and data-driven reasoning. InThe Twelfth International Con- ference on Learning Representations. ICLR, 2023. Source of the Alzheimer’s expert-consen...

    work page 2023

  3. [3]

    DAGMA: learning DAGs via M- matrices and a log-determinant acyclicity characterization

    Kevin Bello, Bryon Aragam, and Pradeep Ravikumar. DAGMA: learning DAGs via M- matrices and a log-determinant acyclicity characterization. InAdvances in Neural Information Processing Systems. Curran Associates, Inc., 2022

    work page 2022

  4. [4]

    Graph of thoughts: solving elaborate problems with large language models.Pro- ceedings of the AAAI Conference on Artificial Intelligence, 38(16):17682–17690, 2024

    Maciej Besta, Nils Blach, Ales Kubicek, Robert Gerstenberger, Michal Podstawski, Lukas Gi- aninazzi, Joanna Gajda, Tomasz Lehmann, Hubert Niewiadomski, Piotr Nyczyk, and Torsten Hoefler. Graph of thoughts: solving elaborate problems with large language models.Pro- ceedings of the AAAI Conference on Artificial Intelligence, 38(16):17682–17690, 2024. doi: 1...

    doi:10.1609/aaai.v38i16.29720 2024

  5. [5]

    Causal discovery for fairness

    R¯ uta Binkyt˙ e, Karima Makhlouf, Carlos Pinz´ on, Sami Zhioua, and Catuscia Palamidessi. Causal discovery for fairness. InWorkshop on Algorithmic Fairness through the Lens of Causal- ity and Privacy, pages 7–22. PMLR, PMLR, 2023

    work page 2023

  6. [6]

    The landscape of causal discovery data: grounding causal discovery in real-world applications, 2025

    Philippe Brouillard, Chandler Squires, Jonas Wahl, Konrad P Kording, Karen Sachs, Alexan- dre Drouin, and Dhanya Sridhar. The landscape of causal discovery data: grounding causal discovery in real-world applications, 2025. arXiv:2412.01953

    arXiv 2025

  7. [7]

    Path-specific counterfactual fairness

    Silvia Chiappa. Path-specific counterfactual fairness. InProceedings of the AAAI Conference on Artificial Intelligence, volume 33, pages 7801–7808. AAAI Press, 2019

    work page 2019

  8. [8]

    Optimal structure identification with greedy search.Journal of Machine Learning Research, 3(Nov):507–554, 2002

    David Maxwell Chickering. Optimal structure identification with greedy search.Journal of Machine Learning Research, 3(Nov):507–554, 2002

    work page 2002

Show all 50 references
  1. [9]

    Order-independent constraint-based causal structure learning.Journal of Machine Learning Research, 15(1):3741–3782, 2014

    Diego Colombo and Marloes H Maathuis. Order-independent constraint-based causal structure learning.Journal of Machine Learning Research, 15(1):3741–3782, 2014

    2014

  2. [10]

    Causal machine learning for predicting treatment outcomes.Nature Medicine, 30:958–968, 2024

    Stefan Feuerriegel, Dennis Frauen, Valentyn Melnychuk, Jonas Schweisthal, Konstantin Hess, Alicia Curth, Stefan Bauer, Niki Kilbertus, Isaac S Kohane, and Mihaela van der Schaar. Causal machine learning for predicting treatment outcomes.Nature Medicine, 30:958–968, 2024

    2024

  3. [11]

    Review of causal discovery methods based on graphical models.Frontiers in Genetics, 10:524, 2019

    Clark Glymour, Kun Zhang, and Peter Spirtes. Review of causal discovery methods based on graphical models.Frontiers in Genetics, 10:524, 2019

    2019

  4. [12]

    Generalized score functions for causal discovery

    Biwei Huang, Kun Zhang, Yizhu Lin, Bernhard Sch¨ olkopf, and Clark Glymour. Generalized score functions for causal discovery. InProceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 1551–1560. ACM, 2018

    2018

  5. [13]

    Efficient causal graph discovery using large language models, 2024

    Thomas Jiralerspong, Xiaoyin Chen, Yash More, Vedant Shah, and Yoshua Bengio. Efficient causal graph discovery using large language models, 2024. arXiv:2402.01207

    2024 arXiv

  6. [14]

    Estimating high-dimensional directed acyclic graphs with the PC-algorithm.Journal of Machine Learning Research, 8:613–636, 2007

    Markus Kalisch and Peter B¨ uhlmann. Estimating high-dimensional directed acyclic graphs with the PC-algorithm.Journal of Machine Learning Research, 8:613–636, 2007

    2007

  7. [15]

    ALCM: autonomous LLM-augmented causal discovery framework, 2024

    Elahe Khatibi, Mahyar Abbasian, Zhongqi Yang, Iman Azimi, and Amir M Rahmani. ALCM: autonomous LLM-augmented causal discovery framework, 2024. arXiv:2405.01744. 16

    2024 arXiv

  8. [16]

    Causal reasoning and large language models: opening a new frontier for causality.Transactions on Machine Learning Research, TMLR, 2023

    Emre Kiciman, Robert Ness, Amit Sharma, and Chenhao Tan. Causal reasoning and large language models: opening a new frontier for causality.Transactions on Machine Learning Research, TMLR, 2023

    2023

  9. [17]

    Avoiding discrimination through causal reasoning

    Niki Kilbertus, Mateo Rojas-Carulla, Giambattista Parascandolo, Moritz Hardt, Dominik Janzing, and Bernhard Sch¨ olkopf. Avoiding discrimination through causal reasoning. InAd- vances in Neural Information Processing Systems. Curran Associates, Inc., 2017

    2017

  10. [18]

    A survey of Bayesian network structure learning.Artificial Intelligence Review, 56:8721–8814, 2023

    Neville K Kitson, Anthony C Constantinou, Zhigao Guo, Yang Liu, and Kiattikun Chobtham. A survey of Bayesian network structure learning.Artificial Intelligence Review, 56:8721–8814, 2023

    2023

  11. [19]

    Counterfactual fairness

    Matt J Kusner, Joshua Loftus, Chris Russell, and Ricardo Silva. Counterfactual fairness. In Advances in Neural Information Processing Systems. Curran Associates, Inc., 2017

    2017

  12. [20]

    Causal discovery with language models as imperfect experts, 2023

    Stephanie Long, Tibor Schuster, and Alexandre Pich´ e. Causal discovery with language models as imperfect experts, 2023. arXiv:2307.02390

    2023 arXiv

  13. [21]

    When causality meets fairness: a survey.Journal of Logical and Algebraic Methods in Programming, 141:101000, 2024

    Karima Makhlouf, Sami Zhioua, and Catuscia Palamidessi. When causality meets fairness: a survey.Journal of Logical and Algebraic Methods in Programming, 141:101000, 2024

    2024

  14. [22]

    Fair inference on outcomes

    Razieh Nabi and Ilya Shpitser. Fair inference on outcomes. InProceedings of the AAAI Conference on Artificial Intelligence. AAAI Press, 2018

    2018

  15. [23]

    Direct and indirect effects.Proceedings of the Seventeenth Conference on Uncer- tainty in Artificial Intelligence, UAI-01:411–420, 2001

    Judea Pearl. Direct and indirect effects.Proceedings of the Seventeenth Conference on Uncer- tainty in Artificial Intelligence, UAI-01:411–420, 2001

    2001

  16. [24]

    Cambridge university press, 2009

    Judea Pearl.Causality. Cambridge university press, 2009

    2009

  17. [25]

    MIT Press, 2017

    Jonas Peters, Dominik Janzing, and Bernhard Sch¨ olkopf.Elements of causal inference: foun- dations and learning algorithms. MIT Press, 2017

    2017

  18. [26]

    Causal fairness analysis: a causal toolkit for fair machine learning.Foundations and Trends in Machine Learning, 17(3):304–589, 2024

    Drago Plecko and Elias Bareinboim. Causal fairness analysis: a causal toolkit for fair machine learning.Foundations and Trends in Machine Learning, 17(3):304–589, 2024

    2024

  19. [27]

    Reconciling predictive and statistical parity: a causal approach

    Drago Plecko and Elias Bareinboim. Reconciling predictive and statistical parity: a causal approach. InProceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 14625–14632. AAAI Press, 2024

    2024

  20. [28]

    Causal fairness analysis: a causal toolkit for fair machine learning.Foundations and Trends in Machine Learning, 17(3):304–589, 2024

    Drago Plecko and Elias Bareinboim. Causal fairness analysis: a causal toolkit for fair machine learning.Foundations and Trends in Machine Learning, 17(3):304–589, 2024. doi: 10.1561/ 2200000106

    2024

  21. [29]

    Fairness-accuracy trade-offs: a causal perspective

    Drago Plecko and Elias Bareinboim. Fairness-accuracy trade-offs: a causal perspective. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 26344–26353. AAAI Press, 2025

    2025

  22. [30]

    Causal inference and counterfactual prediction in machine learning for actionable healthcare.Nature Machine Intelligence, 2:369–375, 2020

    Mattia Prosperi, Yi Guo, Matthew Sperrin, Jennifer S Koopman, Jae S Min, Xing He, Stephen Rich, Mo Wang, Iain E Buchan, and Jiang Bian. Causal inference and counterfactual prediction in machine learning for actionable healthcare.Nature Machine Intelligence, 2:369–375, 2020

    2020

  23. [31]

    Causal fairness under unobserved confounding: a neural sensitivity framework, 2023

    Maresa Schr¨ oder, Dennis Frauen, and Stefan Feuerriegel. Causal fairness under unobserved confounding: a neural sensitivity framework, 2023. arXiv:2311.18460. 17

    2023 arXiv

  24. [32]

    A linear non-Gaussian acyclic model for causal discovery.Journal of Machine Learning Research, 7(10), 2006

    Shohei Shimizu, Patrik O Hoyer, Aapo Hyv¨ arinen, Antti Kerminen, and Michael Jordan. A linear non-Gaussian acyclic model for causal discovery.Journal of Machine Learning Research, 7(10), 2006

    2006

  25. [33]

    DirectLiNGAM: a direct method for learning a linear non-Gaussian structural equation model.Journal of Machine Learning Research, 12:1225–1248, 2011

    Shohei Shimizu, Takanori Inazumi, Yasuhiro Sogawa, Aapo Hyv¨ arinen, Yoshinobu Kawahara, Takashi Washio, Patrik O Hoyer, and Kenneth Bollen. DirectLiNGAM: a direct method for learning a linear non-Gaussian structural equation model.Journal of Machine Learning Research, 12:1225...

    2011

  26. [34]

    A simple approach for finding the globally optimal Bayesian network structure, 2012

    Tomi Silander and Petri Myllymaki. A simple approach for finding the globally optimal Bayesian network structure, 2012. arXiv:1206.6875

    2012 arXiv

  27. [35]

    MIT press, 2000

    Peter Spirtes, Clark N Glymour, and Richard Scheines.Causation, prediction, and search. MIT press, 2000

    2000

  28. [36]

    Causal inference in the presence of latent variables and selection bias, 2013

    Peter L Spirtes, Christopher Meek, and Thomas S Richardson. Causal inference in the presence of latent variables and selection bias, 2013. arXiv:1302.4983

    2013 arXiv

  29. [37]

    Realizing LLMs’ causal potential requires science-grounded, novel benchmarks, 2025

    Ashutosh Srivastava, Lokesh Nagalapatti, Gautam Jajoo, Aniket Vashishtha, Parameswari Kr- ishnamurthy, and Amit Sharma. Realizing LLMs’ causal potential requires science-grounded, novel benchmarks, 2025. arXiv:2510.16530

    2025

  30. [38]

    Causal discovery in the presence of missing data

    Ruibo Tu, Cheng Zhang, Paul Ackermann, Karthika Mohan, Hedvig Kjellstr¨ om, and Kun Zhang. Causal discovery in the presence of missing data. InProceedings of the 22nd Interna- tional Conference on Artificial Intelligence and Statistics, pages 1762–1770. PMLR, PMLR, 2019

    2019

  31. [39]

    D’ya like DAGs? a survey on structure learning and causal discovery.ACM Computing Surveys, 55(4):1–36, 2022

    Matthew J Vowels, Necati Cihan Camgoz, and Richard Bowden. D’ya like DAGs? a survey on structure learning and causal discovery.ACM Computing Surveys, 55(4):1–36, 2022

    2022

  32. [40]

    Chain-of-thought prompting elicits reasoning in large language models

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. Chain-of-thought prompting elicits reasoning in large language models. InAdvances in Neural Information Processing Systems. Curran Associates, Inc., 2022

    2022

  33. [41]

    PC-fairness: a unified framework for measuring causality-based fairness

    Yongkai Wu, Lu Zhang, and Xintao Wu. PC-fairness: a unified framework for measuring causality-based fairness. InAdvances in Neural Information Processing Systems. Curran As- sociates, Inc., 2019

    2019

  34. [42]

    Tree of thoughts: deliberate problem solving with large language models

    Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Thomas L Griffiths, Yuan Cao, and Karthik Narasimhan. Tree of thoughts: deliberate problem solving with large language models. In Advances in Neural Information Processing Systems. Curran Associates, Inc., 2023

    2023

  35. [43]

    DAG-GNN: DAG structure learning with graph neural networks

    Yue Yu, Jie Chen, Tian Gao, and Mo Yu. DAG-GNN: DAG structure learning with graph neural networks. InProceedings of the 36th International Conference on Machine Learning. PMLR, 2019

    2019

  36. [44]

    Learning optimal Bayesian networks: a shortest path perspective.Journal of Artificial Intelligence Research, 48:23–65, 2013

    Changhe Yuan and Brandon Malone. Learning optimal Bayesian networks: a shortest path perspective.Journal of Artificial Intelligence Research, 48:23–65, 2013

    2013

  37. [45]

    Fairness-driven LLM-based causal discovery with active learning and dynamic scoring, 2025

    Khadija Zanna and Akane Sano. Fairness-driven LLM-based causal discovery with active learning and dynamic scoring, 2025. arXiv:2503.17569. 18

    2025 arXiv

  38. [46]

    gCastle: a python toolbox for causal discovery, 2021

    Keli Zhang, Shengyu Zhu, Marcus Kalander, Ignavier Ng, Junjian Ye, Zhitang Chen, and Lujia Pan. gCastle: a python toolbox for causal discovery, 2021

    2021

  39. [47]

    DAGs with NO TEARS: continuous optimization for structure learning

    Xun Zheng, Bryon Aragam, Pradeep Ravikumar, and Eric P Xing. DAGs with NO TEARS: continuous optimization for structure learning. InAdvances in Neural Information Processing Systems. Curran Associates, Inc., 2018

    2018

  40. [48]

    Learning sparse nonparametric DAGs

    Xun Zheng, Chen Dan, Bryon Aragam, Pradeep Ravikumar, and Eric P Xing. Learning sparse nonparametric DAGs. InInternational Conference on Artificial Intelligence and Statistics. PMLR, 2020

    2020

  41. [49]

    Causal-learn: causal discovery in python.Journal of Machine Learning Research, 25(60):1–8, 2024

    Yujia Zheng, Biwei Huang, Wei Chen, Joseph Ramsey, Mingming Gong, Ruichu Cai, Shohei Shimizu, Peter Spirtes, and Kun Zhang. Causal-learn: causal discovery in python.Journal of Machine Learning Research, 25(60):1–8, 2024

    2024

  42. [50]

    Tracing causal paths from experimental and observational data.The Journal of Politics, 85(1):250–265, 2023

    Xiang Zhou and Teppei Yamamoto. Tracing causal paths from experimental and observational data.The Journal of Politics, 85(1):250–265, 2023. 19

    2023

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.