{"id":"https://openalex.org/W7162430530","doi":"https://doi.org/10.48550/arxiv.2605.24733","title":"StepGap: A Hybrid NLI-LLM Checker for Step-Level Evidence-Gap Detectionin Multi-Hop Question Answering","display_name":"StepGap: A Hybrid NLI-LLM Checker for Step-Level Evidence-Gap Detectionin Multi-Hop Question Answering","publication_year":2026,"publication_date":"2026-05-23","ids":{"openalex":"https://openalex.org/W7162430530","doi":"https://doi.org/10.48550/arxiv.2605.24733"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.24733","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.24733","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.24733","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5026482925","display_name":"Yuelyu Ji","orcid":"https://orcid.org/0000-0001-6389-5823"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ji, Yuelyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137017108","display_name":"Zhuochun Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Zhuochun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030046423","display_name":"Hui Ji","orcid":"https://orcid.org/0000-0002-1674-6056"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ji, Hui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137034399","display_name":"Daqing He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Daqing","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/T10028","display_name":"Topic Modeling","score":0.47279998660087585,"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":0.47279998660087585,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.37540000677108765,"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/T13629","display_name":"Text Readability and Simplification","score":0.0649000033736229,"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/question-answering","display_name":"Question answering","score":0.6733999848365784},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.5274999737739563},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5080000162124634},{"id":"https://openalex.org/keywords/interval","display_name":"Interval (graph theory)","score":0.45350000262260437},{"id":"https://openalex.org/keywords/tree","display_name":"Tree (set theory)","score":0.4332999885082245},{"id":"https://openalex.org/keywords/sign","display_name":"Sign (mathematics)","score":0.3785000145435333},{"id":"https://openalex.org/keywords/limit","display_name":"Limit (mathematics)","score":0.34689998626708984},{"id":"https://openalex.org/keywords/confidence-interval","display_name":"Confidence interval","score":0.3255000114440918}],"concepts":[{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.6733999848365784},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6661999821662903},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5400000214576721},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.5274999737739563},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5080000162124634},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4729999899864197},{"id":"https://openalex.org/C2778067643","wikidata":"https://www.wikidata.org/wiki/Q166507","display_name":"Interval (graph theory)","level":2,"score":0.45350000262260437},{"id":"https://openalex.org/C113174947","wikidata":"https://www.wikidata.org/wiki/Q2859736","display_name":"Tree (set theory)","level":2,"score":0.4332999885082245},{"id":"https://openalex.org/C139676723","wikidata":"https://www.wikidata.org/wiki/Q1193832","display_name":"Sign (mathematics)","level":2,"score":0.3785000145435333},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.34689998626708984},{"id":"https://openalex.org/C44249647","wikidata":"https://www.wikidata.org/wiki/Q208498","display_name":"Confidence interval","level":2,"score":0.3255000114440918},{"id":"https://openalex.org/C84525736","wikidata":"https://www.wikidata.org/wiki/Q831366","display_name":"Decision tree","level":2,"score":0.325300008058548},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.30979999899864197},{"id":"https://openalex.org/C146357865","wikidata":"https://www.wikidata.org/wiki/Q1123245","display_name":"Stage (stratigraphy)","level":2,"score":0.3082999885082245},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.29010000824928284},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.27869999408721924},{"id":"https://openalex.org/C2776730729","wikidata":"https://www.wikidata.org/wiki/Q1392130","display_name":"Flag (linear algebra)","level":3,"score":0.2786000072956085},{"id":"https://openalex.org/C59656382","wikidata":"https://www.wikidata.org/wiki/Q191536","display_name":"Conjunction (astronomy)","level":2,"score":0.2754000127315521},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.26260000467300415},{"id":"https://openalex.org/C182306322","wikidata":"https://www.wikidata.org/wiki/Q1779371","display_name":"Order (exchange)","level":2,"score":0.2606000006198883}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.24733","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.24733","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.24733","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.24733","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":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","score":0.7225010991096497,"display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,95],"present":[1],"\\textbf{StepGap},":[2],"a":[3,36,63,83,98,120,142],"hybrid":[4],"NLI-LLM":[5],"decision":[6],"tree":[7],"that":[8,108],"detects":[9],"step-level":[10,113],"evidence":[11],"gaps":[12],"in":[13],"multi-hop":[14,42],"QA":[15],"and":[16],"emits":[17],"one":[18],"of":[19,56,76,85],"three":[20,75,135],"typed":[21,121],"labels:":[22],"\\textsc{Contradicted":[23],"Claim}":[24],"(CC),":[25],"\\textsc{Irrelevant":[26],"Evidence}":[27],"(IE),":[28],"or":[29],"\\textsc{Missing":[30],"Bridge}":[31],"(MB),":[32],"each":[33,92],"tied":[34],"to":[35,132],"concrete":[37],"repair":[38],"action.":[39],"On":[40],"82":[41],"questions":[43],"(181":[44],"annotated":[45],"steps,":[46],"$\u03ba{=}0.704$),":[47],"StepGap":[48,68,125],"reaches":[49],"sF1$=$72.0,":[50],"within":[51],"the":[52,115,138,148],"bootstrap":[53],"confidence":[54],"interval":[55],"an":[57],"LLM-only":[58,78],"baseline":[59],"(70.1)":[60],"but":[61],"with":[62,137],"more":[64],"decomposable":[65],"structure:":[66],"every":[67,110],"stage":[69],"\\emph{hurts}":[70],"F1":[71,81,102,114],"when":[72],"removed,":[73],"while":[74],"four":[77],"removals":[79],"\\emph{improve}":[80],"--":[82],"sign":[84],"\\emph{competing-error":[86],"cancellation},":[87],"where":[88],"internal":[89],"stages":[90],"mask":[91],"other's":[93],"errors.":[94],"further":[96],"expose":[97],"\\emph{Q-F1":[99],"trap}:":[100],"question-level":[101],"is":[103],"mechanically":[104],"inflated":[105],"by":[106],"checkers":[107],"flag":[109],"step,":[111],"making":[112],"necessary":[116],"diagnostic.":[117],"Used":[118],"as":[119],"GRPO":[122,151],"process":[123],"reward,":[124],"improves":[126],"Qwen2.5-7B-Instruct":[127],"Exact":[128],"Match":[129],"from":[130],"$32.1{\\pm}0.3$":[131],"$35.4{\\pm}0.9$":[133],"across":[134],"seeds,":[136],"single-run":[139],"comparison":[140],"showing":[141],"$+5.6$":[143],"Avg":[144],"EM":[145],"gain":[146],"over":[147],"matched":[149],"Search-R1":[150],"reproduction.":[152]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-27T00:00:00"}