ARXIV:2606.03602 · CAUSAL DISCOVERY · SUBMITTED 03 JUN · 20:32 UTC · FRESHNESS FRESH

VerifiedSource: PDF linkedVerifiedPaperPack: citation fields availablePartialProof: unverified proof status

CauTion: Knowing When to Trust LLMs for Ensemble Causal Discovery

Bo Peng · Kaiwen Wu · Sirui Chen · Zhiheng Wang · Yu Qiao · Chaochao Lu · arXiv

A framework that reliably integrates LLM domain knowledge into an ensemble of statistical causal discovery algorithms for improved accuracy and robustness.

Ship in 2-4 weeks›Score7.0Evidence unverified

Opportunity summary

Pain A framework that reliably integrates LLM domain knowledge into an ensemble of statistical causal discovery algorithms for improved accuracy and robustness.

Evidence 0 refs | 4 sources | 83% coverage

Blocker Evidence unverified

Open Build Read PDF Signal Canvas Track

PROBLEM

A framework that reliably integrates LLM domain knowledge into an ensemble of statistical causal discovery algorithms for improved accuracy and robustness. While large language models (LLMs) offer a promising source of domain knowledge to…

METHOD

Full abstract

Causal discovery from observational data remains challenging due to the fundamental limitations of purely statistical methods, such as statistical distinguishability within equivalence classes and sensitivity to finite sample sizes. While large language models (LLMs) offer a promising source of domain knowledge to complement statistical inference, existing LLM-augmented methods are vulnerable to LLM errors and incur high token costs. Moreover, reliance on a single data-centric algorithm can make results sensitive to algorithm-specific biases. To address these limitations, we propose CauTion, a framework that reliably integrates LLM domain knowledge into an ensemble of statistical causal discovery algorithms through consensus filtering and LLM reliability estimation. CauTion proceeds in three stages. First, an algorithm ensemble utilizes a consensus voting to resolve up to 96% of edges on which algorithms agree, achieving near-perfect accuracy on the filtered consensus edges. Second, a trust-calibrated arbitration mechanism estimates the relative reliability of the LLM and the algorithms via an annotation-free trust calibration procedure, which is then utilized to govern a trust-weighted voting process that restricts LLM arbitration exclusively to edges with unreliable algorithmic evidence. Third, a cycle repair step is applied to guarantee the final causal graph is validly acyclic. Experiments on six datasets demonstrate that CauTion consistently outperforms both data-centric and LLM-augmented baselines, with larger gains on larger graphs and strong robustness to LLM errors. Code is available at https://github.com/OpenCausaLab/CauTion.

RESULT

ScienceToStartup currently rates this 7.0/10 on the public viability pass. Moreover, reliance on a single data-centric algorithm can make results sensitive to algorithm-specific biases. A public repository is linked, so build verification can inspect…

WHY NOW

Causal Discovery moved forward this cycle; last verified June 2026. Public score 7.0/10. Implementation evidence is present through a linked repository.

Continue into Read for claims, analysis, references, and neighboring papers.

Opportunity summary

Score7.0

PainA framework that reliably integrates LLM domain knowledge into an ensemble of statistical causal discovery algorithms for improved accuracy and robustness.

Evidence0 refs | 4 sources | 83% coverage

Blockerno shell-level blocker reported

Analysis summary

A framework that reliably integrates LLM domain knowledge into an ensemble of statistical causal discovery algorithms for improved accuracy and robustness.

VerifiedSource: PDF linkedVerifiedPaperPack: citation fields availablePartialProof: unverified proof status

Competitive landscape

A framework that reliably integrates LLM domain knowledge into an ensemble of statistical causal discovery algorithms for improved accuracy and robustness.

Segment

Causal Discovery

Adoption evidence

Public code linked for build inspection

Commercial read

7.0/10 public viability

Direct

not classified

Adjacent

not classified

Substitute

not classified

Unknown

not classified

{ "contract_version": "paper-r2", "paper_id": "b61d1b8b-0fc3-4279-83e9-39f37f1a2a26", "arxiv_id": "2606.03602", "canonical_route": "/paper/caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery", "active_tab": "synced from current hash by the drawer client", "selected_artifact": "caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery", "endpoints": { "paper_pack": "/api/v1/paper/caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery/paper-pack", "build_passport": "/api/v1/paper/caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery/build-passport", "mcp_resource": "sciencetostartup://surfaces/paper-workspace" } }

{ "surface": "paper", "mode": "paper", "query": "CauTion: Knowing When to Trust LLMs for Ensemble Causal Discovery", "normalized_query": "2606.03602", "route": "/paper/caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery", "paper_ref": "caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery", "topic_slug": null, "benchmark_ref": null, "dataset_ref": null }

{ "@context": "https://schema.org", "@graph": [ { "@type": "WebPage", "@id": "https://sciencetostartup.com/paper/caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery#webpage", "url": "https://sciencetostartup.com/paper/caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery", "name": "CauTion: Knowing When to Trust LLMs for Ensemble Causal Discovery", "description": "A framework that reliably integrates LLM domain knowledge into an ensemble of statistical causal discovery algorithms for improved accuracy and robustness.", "isPartOf": { "@id": "https://sciencetostartup.com/#website" } }, { "@type": "ScholarlyArticle", "@id": "https://sciencetostartup.com/paper/caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery#scholarlyArticle", "headline": "CauTion: Knowing When to Trust LLMs for Ensemble Causal Discovery", "description": "A framework that reliably integrates LLM domain knowledge into an ensemble of statistical causal discovery algorithms for improved accuracy and robustness.", "url": "https://sciencetostartup.com/paper/caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery", "sameAs": "https://arxiv.org/abs/2606.03602", "identifier": { "@type": "PropertyValue", "propertyID": "arXiv", "value": "2606.03602" }, "isAccessibleForFree": true, "isPartOf": { "@id": "https://sciencetostartup.com/#website" }, "datePublished": "2026-06-02T13:07:43.000Z", "author": [ { "@type": "Person", "name": "Bo Peng" }, { "@type": "Person", "name": "Kaiwen Wu" }, { "@type": "Person", "name": "Sirui Chen" }, { "@type": "Person", "name": "Zhiheng Wang" }, { "@type": "Person", "name": "Yu Qiao" }, { "@type": "Person", "name": "Chaochao Lu" } ], "codeRepository": "https://github.com/OpenCausaLab/CauTion", "additionalProperty": [ { "@type": "PropertyValue", "propertyID": "viabilityScore", "value": 7 }, { "@type": "PropertyValue", "propertyID": "researchDomain", "value": "Causal Discovery" }, { "@type": "PropertyValue", "propertyID": "commercialReadiness", "value": "code, repo url" } ] }, { "@type": "SoftwareSourceCode", "@id": "https://sciencetostartup.com/paper/caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery#software", "name": "CauTion: Knowing When to Trust LLMs for Ensemble Causal Discovery - Source Code", "description": "A framework that reliably integrates LLM domain knowledge into an ensemble of statistical causal discovery algorithms for improved accuracy and robustness.", "codeRepository": "https://github.com/OpenCausaLab/CauTion", "url": "https://github.com/OpenCausaLab/CauTion" }, { "@type": "BreadcrumbList", "itemListElement": [ { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://sciencetostartup.com" }, { "@type": "ListItem", "position": 2, "name": "Causal Discovery", "item": "https://sciencetostartup.com/topics" }, { "@type": "ListItem", "position": 3, "name": "CauTion: Knowing When to Trust LLMs for Ensemble Causal Disc", "item": "https://sciencetostartup.com/paper/caution-knowing-when-to-trust-llms-for-ensemble-causal-discovery" } ] } ] }

Competitive landscape

A framework that reliably integrates LLM domain knowledge into an ensemble of statistical causal discovery algorithms for improved accuracy and robustness.

Segment

Causal Discovery

Adoption evidence

Public code linked for build inspection

Commercial read

7.0/10 public viability

Direct

not classified

Adjacent

not classified

Substitute

not classified

Unknown

not classified

CauTion: Knowing When to Trust LLMs for Ensemble Causal Discovery

CauTion: Knowing When to Trust LLMs for Ensemble Causal Discovery

Claim map

Constellation map

Competitive landscape

Buzz

PDF

REFERENCES

Related Papers

Related Resources

Subscribe to the weekly brief

Build artifacts

Brief

Experiment plan

Validation checklist

Scientific founder

Translational engineer

Domain operator

GTM lead

Regulatory/clinical advisor

Timeline

Claim map

Constellation map

Competitive landscape

Buzz

PDF

REFERENCES

Related Papers

Related Resources

Subscribe to the weekly brief

Build artifacts

Brief

Experiment plan

Validation checklist

Scientific founder

Translational engineer

Domain operator

GTM lead

Regulatory/clinical advisor

Timeline