{"id":"https://openalex.org/W7123454532","doi":"https://doi.org/10.48550/arxiv.2601.07754","title":"Structure First, Reason Next: Enhancing a Large Language Model using Knowledge Graph for Numerical Reasoning in Financial Documents","display_name":"Structure First, Reason Next: Enhancing a Large Language Model using Knowledge Graph for Numerical Reasoning in Financial Documents","publication_year":2026,"publication_date":"2026-01-12","ids":{"openalex":"https://openalex.org/W7123454532","doi":"https://doi.org/10.48550/arxiv.2601.07754"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2601.07754","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2601.07754","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,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2601.07754","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5122939843","display_name":"Aryan Mishra","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mishra, Aryan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5120689746","display_name":"Akash Anil","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Anil, Akash","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":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/T11326","display_name":"Stock Market Forecasting Methods","score":0.24549999833106995,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T11326","display_name":"Stock Market Forecasting Methods","score":0.24549999833106995,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.1518000066280365,"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":0.07840000092983246,"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/schema","display_name":"Schema (genetic algorithms)","score":0.5936999917030334},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.5917999744415283},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.39879998564720154},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.39739999175071716},{"id":"https://openalex.org/keywords/logical-reasoning","display_name":"Logical reasoning","score":0.3334999978542328},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.3312000036239624},{"id":"https://openalex.org/keywords/financial-modeling","display_name":"Financial modeling","score":0.322299987077713},{"id":"https://openalex.org/keywords/unstructured-data","display_name":"Unstructured data","score":0.303600013256073}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6389999985694885},{"id":"https://openalex.org/C10138342","wikidata":"https://www.wikidata.org/wiki/Q43015","display_name":"Finance","level":1,"score":0.6154999732971191},{"id":"https://openalex.org/C52146309","wikidata":"https://www.wikidata.org/wiki/Q7431116","display_name":"Schema (genetic algorithms)","level":2,"score":0.5936999917030334},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.5917999744415283},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.39879998564720154},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.39739999175071716},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33559998869895935},{"id":"https://openalex.org/C43971567","wikidata":"https://www.wikidata.org/wiki/Q3142865","display_name":"Logical reasoning","level":2,"score":0.3334999978542328},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.3312000036239624},{"id":"https://openalex.org/C23925645","wikidata":"https://www.wikidata.org/wiki/Q5449731","display_name":"Financial modeling","level":2,"score":0.322299987077713},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.31349998712539673},{"id":"https://openalex.org/C2781252014","wikidata":"https://www.wikidata.org/wiki/Q1141900","display_name":"Unstructured data","level":3,"score":0.303600013256073},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2937999963760376},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2881999909877777},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.28290000557899475},{"id":"https://openalex.org/C113336015","wikidata":"https://www.wikidata.org/wiki/Q574010","display_name":"Complete information","level":2,"score":0.27250000834465027},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.27149999141693115},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.26840001344680786},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.26409998536109924},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.2621000111103058},{"id":"https://openalex.org/C192028432","wikidata":"https://www.wikidata.org/wiki/Q845739","display_name":"Query language","level":2,"score":0.26170000433921814},{"id":"https://openalex.org/C161301231","wikidata":"https://www.wikidata.org/wiki/Q3478658","display_name":"Knowledge representation and reasoning","level":2,"score":0.25999999046325684},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.2599000036716461},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.2547999918460846}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2601.07754","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2601.07754","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,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2601.07754","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2601.07754","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,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"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":{"Numerical":[0],"reasoning":[1,109,176],"is":[2,140],"an":[3,141,203],"important":[4,142],"task":[5],"in":[6,14,41,80,110,149],"the":[7,24,47,83,130,188,198,214,225],"analysis":[8],"of":[9,49,59,132],"financial":[10,30,63,72,87,150,157],"documents.":[11],"It":[12],"helps":[13],"understanding":[15],"and":[16,61,95,98],"performing":[17,100],"numerical":[18,90,108,175],"predictions":[19,131,173],"with":[20,46,135,171],"logical":[21,50,136],"conclusions":[22],"for":[23,68,107,155,174],"given":[25],"query":[26],"seeking":[27],"answers":[28],"from":[29,92,187],"texts.":[31],"Recently,":[32],"Large":[33],"Language":[34],"Models":[35],"(LLMs)":[36],"have":[37,116,127],"shown":[38,117],"promising":[39],"results":[40],"multiple":[42],"Question-Answering":[43],"(Q-A)":[44],"systems":[45],"capability":[48],"reasoning.":[51],"As":[52],"documents":[53],"related":[54],"to":[55,144,164,224],"finance":[56],"often":[57,77],"consist":[58],"long":[60],"complex":[62],"contexts,":[64],"LLMs":[65,76,133,154],"appear":[66],"well-suited":[67],"building":[69],"high-quality":[70],"automated":[71],"question-answering":[73],"systems.":[74],"However,":[75],"face":[78],"challenges":[79],"accurately":[81],"processing":[82],"various":[84,156],"numbers":[85],"within":[86],"reports.":[88],"Extracting":[89],"data":[91,120,200],"unstructured":[93],"text":[94],"semi-structured":[96],"tables,":[97],"reliably":[99],"accurate":[101],"calculations,":[102],"remains":[103],"a":[104,162,183],"significant":[105],"bottleneck":[106],"most":[111],"state-of-the-art":[112],"LLMs.":[113],"Recent":[114],"studies":[115],"that":[118,213],"structured":[119,147,166],"augmentations,":[121],"such":[122],"as":[123],"Knowledge":[124],"Graphs":[125],"(KGs),":[126],"notably":[128],"improved":[129,217],"along":[134,170],"explanations.":[137],"Thus,":[138],"it":[139],"requirement":[143],"consider":[145],"inherent":[146],"information":[148,167],"reports":[151],"while":[152],"using":[153,168,182,202],"analytics.":[158],"This":[159],"paper":[160],"proposes":[161],"framework":[163,196,216],"incorporate":[165],"KGs":[169,179],"LLM":[172],"tasks.":[177],"The":[178],"are":[180],"extracted":[181],"proposed":[184,195,215],"schema":[185],"inherently":[186],"document":[189],"under":[190],"processing.":[191],"We":[192,211],"evaluated":[193],"our":[194],"over":[197],"benchmark":[199],"FinQA,":[201],"open-source":[204],"LLM,":[205],"namely":[206],"Llama":[207],"3.1":[208],"8B":[209],"Instruct.":[210],"observed":[212],"execution":[218],"accuracy":[219],"by":[220],"approximately":[221],"12%":[222],"relative":[223],"vanilla":[226],"LLM.":[227]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-01-14T00:00:00"}