mlsanityData to 7 Oct 2026

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

arXiv 2608.02877 ↗First posted 3 Aug 2026cs.LGcited by 0

On the map: LLM reasoning → Causal reasoning in LLMs

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Causal reasoning in LLMs, one of 17 directions in LLM reasoning. Builds on 31 of its 48 references that are in the map. First posted on arXiv 3 Aug 2026.

Builds on

31 of 48 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024ALCM: Autonomous LLM-Augmented Causal Discovery Framework2 May
  2. 2023Causal Reasoning and Large Language Models: Opening a New Frontier for Causality28 Apr

Other references · 29

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,155
  3. 2023Graph of Thoughts: Solving Elaborate Problems with Large Language Modelscited by 873
  4. 2017Counterfactual Fairnesscited by 802
  5. 2018DAGs with NO TEARS: Continuous Optimization for Structure Learningcited by 745
  6. 2019DAG-GNN: DAG Structure Learning with Graph Neural Networkscited by 329
  7. 2013Direct and Indirect Effectscited by 276
  8. 2021D'ya like DAGs? A Survey on Structure Learning and Causal Discoverycited by 209
  9. 2019Learning Sparse Nonparametric DAGscited by 207
  10. 2017Avoiding Discrimination through Causal Reasoningcited by 191
  11. 2018Path-Specific Counterfactual Fairnesscited by 179
  12. 2013Causal Inference in the Presence of Latent Variables and Selection Biascited by 165
17 more
  1. 2017Fair Inference On Outcomescited by 143
  2. 2022DAGMA: Learning DAGs via M-matrices and a Log-Determinant Acyclicity Characterizationcited by 135
  3. 2023Causal-learn: Causal Discovery in Pythoncited by 108
  4. 2021A survey of Bayesian Network structure learningcited by 85
  5. 2024Causal machine learning for predicting treatment outcomescited by 73
  6. 2019PC-Fairness: A Unified Framework for Measuring Causality-based Fairnesscited by 67
  7. 2023Causal Discovery with Language Models as Imperfect Expertscited by 62
  8. 2021gCastle: A Python Toolbox for Causal Discoverycited by 60
  9. 2024Efficient Causal Graph Discovery Using Large Language Modelscited by 51
  10. 2018Causal Discovery in the Presence of Missing Datacited by 46
  11. 2024The Landscape of Causal Discovery Data: Grounding Causal Discovery in Real-World Applicationscited by 17
  12. 2022Causal Discovery for Fairnesscited by 16
  13. 2023Causal Fairness under Unobserved Confounding: A Neural Sensitivity Frameworkcited by 10
  14. 2024Fairness-Accuracy Trade-Offs: A Causal Perspectivecited by 10
  15. 2023Reconciling Predictive and Statistical Parity: A Causal Approachcited by 7
  16. 2025Fairness-Driven LLM-based Causal Discovery with Active Learning and Dynamic Scoringcited by 5
  17. 2025Realizing LLMs' Causal Potential Requires Science-Grounded, Novel Benchmarkscited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026CausalArena: Benchmarking Causal Discovery in the Foundation Model Era
  2. 2026Bypassing the Rationale: Causal Auditing of Implicit Reasoning in Language Models
  3. 2026Causal-Audit: Explicit and Auditable Graph-based Reasoning via Target-Aware Causal Chain Construction
  4. 2025Causal Reasoning in Pieces: Modular In-Context Learning for Causal Discovery
  5. 2024Context-Aware Reasoning On Parametric Knowledge for Inferring Causal Variables
  6. 2026Leveraging Large Language Models for Causal Discovery: a Constraint-based, Argumentation-driven Approach
  7. 2024CausalGraph2LLM: Evaluating LLMs for Causal Queries
  8. 2026RAudit: A Blind Auditing Protocol for Large Language Model Reasoning