{"id":"https://openalex.org/W7214799417","doi":"https://doi.org/10.48550/arxiv.2609.31675","title":"Active Causal Discovery Benchmark: Evaluating LLM Agents Under Budgeted Interventions","display_name":"Active Causal Discovery Benchmark: Evaluating LLM Agents Under Budgeted Interventions","publication_year":2026,"publication_date":"2026-09-16","ids":{"openalex":"https://openalex.org/W7214799417","doi":"https://doi.org/10.48550/arxiv.2609.31675"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.31675","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.31675","pdf_url":"https://arxiv.org/pdf/2609.31675","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"parent_works":[],"lineage":["https://openalex.org/W7214799417"],"child_works_count":0,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.31675","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5141036491","display_name":"Sagar Deb","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Deb, Sagar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109447435","display_name":"Devam Shah","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shah, Devam","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5120250599","display_name":"Ashwanth Krishnan","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Krishnan, Ashwanth","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11303","display_name":"Bayesian Modeling and Causal Inference","score":0.9980000257492065,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11303","display_name":"Bayesian Modeling and Causal Inference","score":0.9980000257492065,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11010","display_name":"Logic, Reasoning, and Knowledge","score":0.0005000000237487257,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12072","display_name":"Machine Learning and Algorithms","score":0.00019999999494757503,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/benchmark","display_name":"benchmark","score":0.5099999904632568},{"id":"https://openalex.org/keywords/llm-based-agents","display_name":"LLM agents","score":0.8140000104904175},{"id":"https://openalex.org/keywords/structural-causal-models","display_name":"structural causal models","score":0.7419999837875366},{"id":"https://openalex.org/keywords/causal-discovery","display_name":"causal discovery","score":0.6629999876022339},{"id":"https://openalex.org/keywords/intervention-efficiency","display_name":"intervention efficiency","score":0.7480000257492065},{"id":"https://openalex.org/keywords/pc-algorithm","display_name":"PC algorithm","score":0.8479999899864197}],"concepts":[{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5051000118255615},{"id":"https://openalex.org/C27415008","wikidata":"https://www.wikidata.org/wiki/Q7256382","display_name":"Psychological intervention","level":2,"score":0.501800000667572},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.47780001163482666},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.4350000023841858},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4212999939918518},{"id":"https://openalex.org/C2780665704","wikidata":"https://www.wikidata.org/wiki/Q959298","display_name":"Intervention (counseling)","level":2,"score":0.4083000123500824},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.3889999985694885},{"id":"https://openalex.org/C74197172","wikidata":"https://www.wikidata.org/wiki/Q1195339","display_name":"Directed acyclic graph","level":2,"score":0.3756999969482422},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37139999866485596},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.3619999885559082},{"id":"https://openalex.org/C51823790","wikidata":"https://www.wikidata.org/wiki/Q504353","display_name":"Greedy algorithm","level":2,"score":0.3409999907016754},{"id":"https://openalex.org/C199521495","wikidata":"https://www.wikidata.org/wiki/Q181487","display_name":"Audit","level":2,"score":0.3248000144958496},{"id":"https://openalex.org/C2780992000","wikidata":"https://www.wikidata.org/wiki/Q17016113","display_name":"Generator (circuit theory)","level":3,"score":0.3172000050544739},{"id":"https://openalex.org/C132964779","wikidata":"https://www.wikidata.org/wiki/Q2110223","display_name":"Raw data","level":2,"score":0.31029999256134033},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.2978000044822693},{"id":"https://openalex.org/C11671645","wikidata":"https://www.wikidata.org/wiki/Q5054567","display_name":"Causal model","level":2,"score":0.2955000102519989},{"id":"https://openalex.org/C173633133","wikidata":"https://www.wikidata.org/wiki/Q2110223","display_name":"Raw score","level":3,"score":0.2921000123023987},{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.2863999903202057},{"id":"https://openalex.org/C2776187449","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Natural language generation","level":3,"score":0.2808000147342682},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.27079999446868896},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.2621999979019165},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.257999986410141},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.2533000111579895}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.31675","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.31675","pdf_url":"https://arxiv.org/pdf/2609.31675","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.31675","doi":"https://doi.org/10.48550/arxiv.2609.31675","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.31675","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.31675","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.31675","pdf_url":"https://arxiv.org/pdf/2609.31675","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7214799417.pdf","grobid_xml":"https://content.openalex.org/works/W7214799417.grobid-xml"},"referenced_works_count":8,"referenced_works":["https://openalex.org/W1568438506","https://openalex.org/W4323927057","https://openalex.org/W4367692219","https://openalex.org/W4385682136","https://openalex.org/W4387561453","https://openalex.org/W4388964563","https://openalex.org/W4415797693","https://openalex.org/W7133216625"],"related_works":["https://openalex.org/W7162403207","https://openalex.org/W7222932463","https://openalex.org/W7162630989","https://openalex.org/W7204793202","https://openalex.org/W7130528626","https://openalex.org/W4398796015","https://openalex.org/W4403362713","https://openalex.org/W4416034908","https://openalex.org/W4415954367","https://openalex.org/W7130543099"],"abstract_inverted_index":{"We":[0],"introduce":[1],"the":[2,52,64,93,139],"Active":[3],"Causal":[4],"Discovery":[5],"Benchmark":[6],"(ACDB),":[7],"an":[8],"SCM-grounded":[9],"environment":[10],"for":[11],"evaluating":[12],"whether":[13],"LLM":[14],"agents":[15],"recover":[16],"causal":[17,165],"graph":[18],"structure":[19],"from":[20],"observations":[21],"and":[22,37,48,82,106,154],"budget-constrained":[23],"hard":[24],"interventions.":[25],"ACDB":[26],"pairs":[27],"a":[28,33,38,58,130,151],"linear-Gaussian":[29],"world":[30],"generator":[31],"with":[32,57,98,103],"fixed":[34],"observe-intervene-submit":[35],"API":[36],"three-layer":[39],"scoring":[40],"contract":[41],"that":[42,160],"separates":[43],"skeleton":[44],"recovery,":[45,47],"DAG":[46,119],"intervention":[49],"efficiency.":[50],"On":[51],"current":[53,144,161],"six-level":[54],"ladder,":[55],"PC":[56,96],"greedy":[59],"active":[60,79,85,164],"orientation":[61],"heuristic":[62],"is":[63,92],"strongest":[65],"non-oracle":[66],"method":[67],"(directed":[68],"F1":[69,124],"42.7%,":[70],"SHD":[71],"4.79),":[72],"ahead":[73],"of":[74],"Claude":[75],"Sonnet":[76],"4.6":[77],"raw":[78,84],"(31.7%,":[80],"7.25)":[81],"GPT-5.4":[83],"(22.9%,":[86],"9.27).":[87],"The":[88,143],"most":[89],"informative":[90],"diagnostic":[91],"precision-recall":[94],"decomposition:":[95],"under-commits":[97],"high":[99],"precision,":[100,105],"LLMs":[101,162],"over-commit":[102],"lower":[104],"statistical-tool":[107],"access":[108],"often":[109],"increases":[110],"abstention":[111],"rather":[112],"than":[113],"useful":[114],"intervention.":[115],"A":[116],"structure-blind":[117],"random":[118],"baseline":[120],"reaches":[121],"23.6%":[122],"directed":[123],"on":[125],"this":[126,134],"dense":[127],"v0":[128],"ladder;":[129],"density":[131],"probe":[132],"lowers":[133],"floor":[135],"to":[136],"16.9%,":[137],"motivating":[138],"v1":[140],"calibration":[141,155],"pass.":[142],"results":[145],"should":[146],"therefore":[147],"be":[148],"read":[149],"as":[150,158],"benchmark":[152],"audit":[153],"report,":[156],"not":[157],"evidence":[159],"solve":[163],"discovery.":[166]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2026-09-30T00:00:00"}
