{"id":"https://openalex.org/W7147297489","doi":"https://doi.org/10.48550/arxiv.2603.28376","title":"Marco DeepResearch: Unlocking Efficient Deep Research Agents via Verification-Centric Design","display_name":"Marco DeepResearch: Unlocking Efficient Deep Research Agents via Verification-Centric Design","publication_year":2026,"publication_date":"2026-03-30","ids":{"openalex":"https://openalex.org/W7147297489","doi":"https://doi.org/10.48550/arxiv.2603.28376"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.28376","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.28376","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.2603.28376","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5132609127","display_name":"Bin Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Bin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132697354","display_name":"Qianghuai Jia","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jia, Qianghuai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132694632","display_name":"Tian Lan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lan, Tian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132707416","display_name":"Junyang Ren","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ren, Junyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132586192","display_name":"Feng Gu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gu, Feng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023498000","display_name":"Feihu Jiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Feihu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132614465","display_name":"Longyue Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Longyue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132676432","display_name":"Zhao Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Zhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5132652637","display_name":"Weihua Luo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luo, Weihua","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":0.4074999988079071,"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":0.4074999988079071,"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/T11948","display_name":"Machine Learning in Materials Science","score":0.13689999282360077,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.06549999862909317,"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/bottleneck","display_name":"Bottleneck","score":0.7397000193595886},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6050999760627747},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.45019999146461487},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.41269999742507935},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.38429999351501465},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.3781999945640564},{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.33559998869895935}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7549999952316284},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.7397000193595886},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6050999760627747},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.526199996471405},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.45019999146461487},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.41269999742507935},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.38429999351501465},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.3781999945640564},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3467999994754791},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.33559998869895935},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3292999863624573},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.31679999828338623},{"id":"https://openalex.org/C111498074","wikidata":"https://www.wikidata.org/wiki/Q173326","display_name":"Formal verification","level":2,"score":0.30559998750686646},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.27000001072883606},{"id":"https://openalex.org/C41550386","wikidata":"https://www.wikidata.org/wiki/Q529909","display_name":"Multi-agent system","level":2,"score":0.2655999958515167},{"id":"https://openalex.org/C75684735","wikidata":"https://www.wikidata.org/wiki/Q858810","display_name":"Big data","level":2,"score":0.2623000144958496},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.25929999351501465},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2524000108242035}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.28376","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.28376","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.2603.28376","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.28376","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":[{"score":0.42807263135910034,"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Deep":[0],"research":[1,82,169],"agents":[2,170],"autonomously":[3],"conduct":[4],"open-ended":[5],"investigations,":[6],"integrating":[7],"complex":[8],"information":[9],"retrieval":[10],"with":[11,85],"multi-step":[12],"reasoning":[13],"across":[14],"diverse":[15],"sources":[16],"to":[17,100,106],"solve":[18],"real-world":[19],"problems.":[20],"To":[21,73],"sustain":[22],"this":[23],"capability":[24],"on":[25,152,171],"long-horizon":[26],"tasks,":[27],"reliable":[28],"verification":[29,49,98,129],"is":[30],"critical":[31],"during":[32],"both":[33],"training":[34,132],"and":[35,57,67,102,115,134,148,178],"inference.":[36],"A":[37],"major":[38],"bottleneck":[39],"in":[40,51],"existing":[41],"paradigms":[42],"stems":[43],"from":[44],"the":[45,69],"lack":[46],"of":[47,185],"explicit":[48,128],"mechanisms":[50,99],"QA":[52,104],"data":[53],"synthesis,":[54],"trajectory":[55,123],"construction,":[56],"test-time":[58],"scaling.":[59],"Errors":[60],"introduced":[61],"at":[62,90,145],"each":[63],"stage":[64],"propagate":[65],"downstream":[66],"degrade":[68],"overall":[70],"agent":[71,83,164],"performance.":[72],"address":[74],"this,":[75],"we":[76],"present":[77],"Marco":[78,139,162,189],"DeepResearch,":[79],"a":[80,86,121,143,182],"deep":[81,168],"optimized":[84],"verification-centric":[87],"framework":[88],"design":[89,120],"three":[91],"levels:":[92],"\\textbf{(1)~QA":[93],"Data":[94],"Synthesis:}":[95],"We":[96,119,137],"introduce":[97],"graph-based":[101],"agent-based":[103],"synthesis":[105,124],"control":[107],"question":[108],"difficulty":[109],"while":[110],"ensuring":[111],"answers":[112],"are":[113],"unique":[114],"correct;":[116],"\\textbf{(2)~Trajectory":[117],"Construction:}":[118],"verification-driven":[122],"method":[125],"that":[126,159],"injects":[127],"patterns":[130],"into":[131],"trajectories;":[133],"\\textbf{(3)~Test-time":[135],"scaling:}":[136],"use":[138],"DeepResearch":[140,163,190],"itself":[141],"as":[142,176],"verifier":[144],"inference":[146],"time":[147],"effectively":[149],"improve":[150],"performance":[151],"challenging":[153,173],"questions.":[154],"Extensive":[155],"experimental":[156],"results":[157],"demonstrate":[158],"our":[160],"proposed":[161],"significantly":[165],"outperforms":[166],"8B-scale":[167],"most":[172],"benchmarks,":[174],"such":[175],"BrowseComp":[177],"BrowseComp-ZH.":[179],"Crucially,":[180],"under":[181],"maximum":[183],"budget":[184],"600":[186],"tool":[187],"calls,":[188],"even":[191],"surpasses":[192],"or":[193],"approaches":[194],"several":[195],"30B-scale":[196],"agents,":[197],"like":[198],"Tongyi":[199],"DeepResearch-30B.":[200]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-02T00:00:00"}