{"id":"https://openalex.org/W7164041497","doi":"https://doi.org/10.48550/arxiv.2606.09748","title":"Multi-Turn Evaluation of Deep Research Agents Under Process-Level Feedback","display_name":"Multi-Turn Evaluation of Deep Research Agents Under Process-Level Feedback","publication_year":2026,"publication_date":"2026-06-08","ids":{"openalex":"https://openalex.org/W7164041497","doi":"https://doi.org/10.48550/arxiv.2606.09748"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.09748","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.09748","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.2606.09748","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138214164","display_name":"Rishabh Sabharwal","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sabharwal, Rishabh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138285503","display_name":"Hongru Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Hongru","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136871004","display_name":"Amos Storkey","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Storkey, Amos","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138285823","display_name":"Jeff Z. Pan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pan, Jeff Z.","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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.19840000569820404,"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"}},"topics":[{"id":"https://openalex.org/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.19840000569820404,"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"}},{"id":"https://openalex.org/T11986","display_name":"Scientific Computing and Data Management","score":0.06390000134706497,"subfield":{"id":"https://openalex.org/subfields/1802","display_name":"Information Systems and Management"},"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/T11948","display_name":"Machine Learning in Materials Science","score":0.06129999831318855,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/rubric","display_name":"Rubric","score":0.822700023651123},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.6499000191688538},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4814999997615814},{"id":"https://openalex.org/keywords/rewriting","display_name":"Rewriting","score":0.45249998569488525},{"id":"https://openalex.org/keywords/sequential-analysis","display_name":"Sequential analysis","score":0.3409000039100647}],"concepts":[{"id":"https://openalex.org/C111640148","wikidata":"https://www.wikidata.org/wiki/Q847349","display_name":"Rubric","level":2,"score":0.822700023651123},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6744999885559082},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.6499000191688538},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4814999997615814},{"id":"https://openalex.org/C154690210","wikidata":"https://www.wikidata.org/wiki/Q1668499","display_name":"Rewriting","level":2,"score":0.45249998569488525},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3610000014305115},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35850000381469727},{"id":"https://openalex.org/C80478641","wikidata":"https://www.wikidata.org/wiki/Q195771","display_name":"Sequential analysis","level":2,"score":0.3409000039100647},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.2955999970436096},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.27720001339912415},{"id":"https://openalex.org/C86426650","wikidata":"https://www.wikidata.org/wiki/Q7452504","display_name":"Sequential estimation","level":2,"score":0.25440001487731934},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.25110000371932983}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.09748","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.09748","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.2606.09748","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.09748","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Existing":[0],"benchmarks":[1],"for":[2,180],"deep":[3],"research":[4,64],"agents":[5,100,150],"(DRAs)":[6],"assess":[7],"only":[8],"single-shot":[9],"outputs,":[10],"ignoring":[11],"a":[12,29,76,116,135],"key":[13,95],"question:":[14],"can":[15],"DRAs":[16,33],"improve":[17],"their":[18],"reports":[19],"when":[20,160],"guided":[21],"by":[22,129],"feedback?":[23],"To":[24,66],"investigate":[25],"this,":[26],"we":[27,70,184],"conduct":[28],"multi-turn":[30,174],"evaluation":[31],"of":[32,81,119,156,178],"under":[34,98],"two":[35],"feedback":[36,121],"settings:":[37],"self-reflection,":[38,99],"in":[39,54,62],"which":[40,55],"the":[41,56,126,162,181],"agent":[42,57],"revises":[43],"its":[44,63],"report":[45,164],"without":[46],"any":[47],"external":[48],"diagnostic":[49],"signal,":[50],"and":[51,83,102,133,188],"process-level":[52,68,120],"feedback,":[53,69],"receives":[58],"guidance":[59],"targeting":[60],"gaps":[61],"strategy.":[65],"enable":[67],"design":[71],"Research":[72],"Gap":[73],"Inference":[74],"(RGI),":[75],"method":[77],"that":[78],"analyzes":[79],"patterns":[80],"satisfied":[82,158],"unsatisfied":[84],"rubric":[85,105],"criteria":[86,106,159],"to":[87,154,165],"infer":[88],"research-process":[89],"gaps.":[90,168],"Our":[91,186],"analysis":[92],"reveals":[93],"three":[94],"findings:":[96],"(i)":[97],"incorporate":[101],"regress":[103,151],"on":[104,152],"at":[107,193],"nearly":[108],"equal":[109],"rates,":[110],"yielding":[111,134],"negligible":[112],"net":[113],"improvement;":[114],"(ii)":[115],"single":[117],"round":[118],"yields":[122],"substantial":[123],"gains,":[124],"raising":[125],"normalized":[127],"score":[128],"approximately":[130],"$8$-$15$":[131],"points":[132],"roughly":[136],"$35$-$40\\%$":[137],"incorporation":[138],"rate;":[139],"(iii)":[140],"these":[141],"gains":[142],"do":[143],"not":[144],"compound":[145],"over":[146],"subsequent":[147],"turns,":[148],"as":[149],"up":[153],"$24\\%$":[155],"previously":[157],"rewriting":[161],"full":[163],"address":[166],"remaining":[167],"Even":[169],"with":[170],"targeted":[171],"guidance,":[172],"reliable":[173],"improvement":[175],"remains":[176],"out":[177],"reach":[179],"DRA":[182],"architectures":[183],"evaluate.":[185],"code":[187],"results":[189],"are":[190],"publicly":[191],"available":[192],"https://github.com/sabharwalrishabh/Multi-Turn-Evaluation-of-DRAs.":[194]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-10T00:00:00"}