{"id":"https://openalex.org/W7162631697","doi":"https://doi.org/10.48550/arxiv.2605.28070","title":"Bridging the Detection-to-Abstention Gap in Reasoning Models under Insufficient Information","display_name":"Bridging the Detection-to-Abstention Gap in Reasoning Models under Insufficient Information","publication_year":2026,"publication_date":"2026-05-27","ids":{"openalex":"https://openalex.org/W7162631697","doi":"https://doi.org/10.48550/arxiv.2605.28070"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.28070","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28070","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.2605.28070","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137221684","display_name":"Renjie Gu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gu, Renjie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137237866","display_name":"Jiaxu Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Jiaxu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137200858","display_name":"Yihao Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yihao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137193372","display_name":"Yun Yue","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yue, Yun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120811888","display_name":"Hansong Xiao","orcid":"https://orcid.org/0009-0005-8555-5543"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao, Hansong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137205796","display_name":"Yefei Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yefei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137200461","display_name":"Yuan Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068097467","display_name":"Chunxiao Guo","orcid":"https://orcid.org/0000-0001-5213-9870"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Chunxiao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063190135","display_name":"Wei Pei","orcid":"https://orcid.org/0000-0002-7367-4618"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Pei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137263497","display_name":"Jinjie Gu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gu, Jinjie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137303811","display_name":"Yixin Cao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cao, Yixin","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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.26179999113082886,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.26179999113082886,"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/T10028","display_name":"Topic Modeling","score":0.21950000524520874,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.11789999902248383,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/bridging","display_name":"Bridging (networking)","score":0.7926999926567078},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6316999793052673},{"id":"https://openalex.org/keywords/deliberation","display_name":"Deliberation","score":0.5163999795913696},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.508899986743927},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.47029998898506165},{"id":"https://openalex.org/keywords/complete-information","display_name":"Complete information","score":0.4478999972343445},{"id":"https://openalex.org/keywords/automated-reasoning","display_name":"Automated reasoning","score":0.4180999994277954},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.40790000557899475}],"concepts":[{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.7926999926567078},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6516000032424927},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6316999793052673},{"id":"https://openalex.org/C2776946740","wikidata":"https://www.wikidata.org/wiki/Q358652","display_name":"Deliberation","level":3,"score":0.5163999795913696},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.508899986743927},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5056999921798706},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.47029998898506165},{"id":"https://openalex.org/C113336015","wikidata":"https://www.wikidata.org/wiki/Q574010","display_name":"Complete information","level":2,"score":0.4478999972343445},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.4180999994277954},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.40790000557899475},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.38179999589920044},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.37709999084472656},{"id":"https://openalex.org/C97364631","wikidata":"https://www.wikidata.org/wiki/Q484284","display_name":"Deductive reasoning","level":2,"score":0.3488999903202057},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.3328999876976013},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.3278000056743622},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3208000063896179},{"id":"https://openalex.org/C190839683","wikidata":"https://www.wikidata.org/wiki/Q2448197","display_name":"Train","level":2,"score":0.3070000112056732},{"id":"https://openalex.org/C73301696","wikidata":"https://www.wikidata.org/wiki/Q5469984","display_name":"Formalism (music)","level":3,"score":0.3003000020980835},{"id":"https://openalex.org/C134752490","wikidata":"https://www.wikidata.org/wiki/Q374182","display_name":"Logical consequence","level":2,"score":0.28790000081062317},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.26980000734329224},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.2599000036716461}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.28070","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28070","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.2605.28070","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28070","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":[{"display_name":"Peace, Justice and strong institutions","score":0.7692599296569824,"id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,34,128,204],"highlight":[1],"a":[2,18,83,104,111,231],"failure":[3],"mode":[4],"of":[5,32,234],"large":[6],"reasoning":[7,25,149,192,212,235,239],"models":[8,14,89,150,168,240],"on":[9,66,124,145,177,214],"questions":[10],"with":[11,139],"insufficient":[12,228],"information:":[13],"may":[15],"recognize":[16],"that":[17,87,152,167,178,207,225],"problem":[19],"is":[20,53,230],"under-specified,":[21],"yet":[22],"still":[23],"continue":[24],"and":[26,135,141,147,160,193,242],"produce":[27],"unsupported":[28,202],"final":[29,49],"answers":[30,64],"instead":[31],"abstaining.":[33],"formalize":[35],"this":[36,77,130],"mismatch":[37],"as":[38,60,103,110],"the":[39,114,186],"detection-to-abstention":[40],"gap,":[41,78],"where":[42,63],"detected":[43],"insufficiency":[44],"fails":[45],"to":[46,90,118,164],"translate":[47],"into":[48],"abstention.":[50],"This":[51],"gap":[52],"especially":[54],"concerning":[55],"in":[56],"high-risk":[57],"domains":[58],"such":[59],"medical":[61],"AI,":[62],"based":[65,123],"incomplete":[67],"evidence":[68],"can":[69,210],"be":[70],"more":[71],"harmful":[72],"than":[73,100],"refusal.":[74],"To":[75],"close":[76],"we":[79],"propose":[80],"Judge-Then-Solve":[81],"(JTS),":[82],"trajectory-level":[84],"reasoning-control":[85],"framework":[86],"trains":[88],"make":[91],"an":[92],"explicit":[93],"answerability":[94,126,187],"commitment":[95],"before":[96],"solution":[97],"generation.":[98],"Rather":[99],"treating":[101],"abstention":[102,157,226],"final-answer":[105],"style,":[106],"JTS":[107,153,189],"casts":[108],"it":[109],"control":[112,236],"decision:":[113],"model":[115],"either":[116],"proceeds":[117],"solve":[119],"or":[120],"terminates":[121],"early":[122],"its":[125],"judgment.":[127],"instantiate":[129],"policy":[131],"through":[132],"supervised":[133],"warm-up":[134],"missing-premise":[136,208],"reinforcement":[137],"learning":[138],"consistency":[140],"length-shaping":[142],"rewards.":[143],"Experiments":[144],"dense":[146],"MoE":[148],"show":[151],"substantially":[154],"improves":[155,194],"reliable":[156],"across":[158],"datasets":[159],"pushes":[161],"Abstention@Detection":[162],"(A@D)":[163],"near-saturation,":[165],"indicating":[166],"not":[169],"only":[170],"detect":[171],"missing":[172],"information":[173,229],"but":[174,216],"also":[175,205],"act":[176],"detection.":[179],"By":[180],"terminating":[181],"unanswerable":[182],"trajectories":[183],"immediately":[184],"after":[185],"judgment,":[188],"reduces":[190],"unnecessary":[191],"inference":[195],"efficiency":[196],"when":[197],"continued":[198],"deliberation":[199],"would":[200],"amplify":[201],"assumptions.":[203],"observe":[206],"training":[209],"alter":[211],"behavior":[213],"difficult":[215],"answerable":[217],"problems,":[218],"reducing":[219],"unproductive":[220],"self-reflection.":[221],"These":[222],"results":[223],"suggest":[224],"under":[227],"key":[232],"form":[233],"for":[237],"deploying":[238],"safely":[241],"efficiently.":[243]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-29T00:00:00"}