{"id":"https://openalex.org/W7161957153","doi":"https://doi.org/10.48550/arxiv.2605.21347","title":"Insights Generator: Systematic Corpus-Level Trace Diagnostics for LLM Agents","display_name":"Insights Generator: Systematic Corpus-Level Trace Diagnostics for LLM Agents","publication_year":2026,"publication_date":"2026-05-20","ids":{"openalex":"https://openalex.org/W7161957153","doi":"https://doi.org/10.48550/arxiv.2605.21347"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.21347","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21347","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.21347","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5115733343","display_name":"Akshay Manglik","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Manglik, Akshay","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114425389","display_name":"Apaar Shanker","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shanker, Apaar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136236716","display_name":"Kaustubh Deshpande","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Deshpande, Kaustubh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136642474","display_name":"Jason Qin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qin, Jason","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136721304","display_name":"Yash Maurya","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Maurya, Yash","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136693570","display_name":"Veronica Chatrath","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chatrath, Veronica","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136684417","display_name":"Vijay S. Kalmath","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kalmath, Vijay S.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136656366","display_name":"Levi Lentz","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lentz, Levi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136676119","display_name":"Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xue, Yuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":9,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"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/T10028","display_name":"Topic Modeling","score":0.25029999017715454,"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/T10028","display_name":"Topic Modeling","score":0.25029999017715454,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.1607999950647354,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.04479999840259552,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/trace","display_name":"TRACE (psycholinguistics)","score":0.7372999787330627},{"id":"https://openalex.org/keywords/generator","display_name":"Generator (circuit theory)","score":0.46779999136924744},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.4602000117301941},{"id":"https://openalex.org/keywords/coding","display_name":"Coding (social sciences)","score":0.43700000643730164},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.3287000060081482},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.32749998569488525}],"concepts":[{"id":"https://openalex.org/C75291252","wikidata":"https://www.wikidata.org/wiki/Q1315756","display_name":"TRACE (psycholinguistics)","level":2,"score":0.7372999787330627},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6978999972343445},{"id":"https://openalex.org/C2780992000","wikidata":"https://www.wikidata.org/wiki/Q17016113","display_name":"Generator (circuit theory)","level":3,"score":0.46779999136924744},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4602000117301941},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.43700000643730164},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3797999918460846},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.37610000371932983},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3571999967098236},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.3287000060081482},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.32749998569488525},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.30300000309944153},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2994999885559082},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.27489998936653137},{"id":"https://openalex.org/C3018587665","wikidata":"https://www.wikidata.org/wiki/Q7268696","display_name":"Qualitative analysis","level":3,"score":0.2732999920845032},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.2718000113964081},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.250900000333786}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.21347","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21347","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.21347","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.21347","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Diagnosing":[0],"failures":[1],"in":[2,165],"LLM":[3],"agents":[4,148],"remains":[5],"largely":[6],"manual.":[7],"Practitioners":[8],"inspect":[9],"a":[10,56,88],"small":[11],"subset":[12],"of":[13,43,45,51,58],"execution":[14,59],"traces,":[15,60],"form":[16],"ad-hoc":[17],"hypotheses,":[18],"and":[19,31,97,115,122,146,154,180],"iterate.":[20],"This":[21],"process":[22],"misses":[23],"patterns":[24,73],"that":[25,69,91],"only":[26],"emerge":[27],"across":[28,74,100,113],"trace":[29,53,75,102],"populations":[30],"does":[32],"not":[33],"scale":[34],"to":[35,64,79,104,168],"production":[36],"corpora":[37],"where":[38],"individual":[39],"traces":[40],"span":[41],"tens":[42],"thousands":[44],"tokens.":[46],"We":[47,82,110],"formalize":[48],"the":[49,61,84,101,142],"problem":[50],"corpus-level":[52],"diagnostics.":[54],"Given":[55],"corpus":[57,103],"goal":[62],"is":[63],"produce":[65,105],"grounded":[66],"natural-language":[67],"insights":[68,108,151],"characterize":[70],"systematic":[71],"behavioral":[72],"groups,":[76],"each":[77],"linked":[78],"supporting":[80],"evidence.":[81],"present":[83],"Insights":[85],"Generator":[86],"(IG),":[87],"multi-agent":[89],"system":[90],"answers":[92],"diagnostic":[93],"questions":[94],"by":[95,127,139],"proposing":[96],"testing":[98],"hypotheses":[99],"an":[106],"evidence-backed":[107],"report.":[109],"evaluate":[111],"IG":[112,129,134,175],"qualitative":[114],"objective":[116],"dimensions,":[117],"spanning":[118],"rubric-based":[119],"report":[120],"assessment":[121],"downstream":[123],"performance":[124,138],"improvements":[125],"achieved":[126],"implementing":[128],"insights.":[130],"Human":[131],"experts":[132,173],"using":[133],"reports":[135,176],"improve":[136],"scaffold":[137],"30.4pp":[140],"over":[141],"unmodified":[143],"baseline":[144],"scaffold,":[145],"coding":[147],"leveraging":[149],"IG-derived":[150],"show":[152],"consistent":[153],"stable":[155],"gains.":[156],"Across":[157],"benchmarks,":[158],"IG's":[159],"scout-investigator":[160],"architecture":[161],"produces":[162],"findings":[163],"comparable":[164],"detection":[166],"coverage":[167],"competing":[169],"approaches,":[170],"while":[171],"domain":[172],"rated":[174],"as":[177],"leading":[178],"depth":[179],"evidence":[181],"quality.":[182]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-22T00:00:00"}