{"id":"https://openalex.org/W7204603595","doi":"https://doi.org/10.48550/arxiv.2608.26218","title":"Same Model, Different Harness: Different Coding-Agent Results","display_name":"Same Model, Different Harness: Different Coding-Agent Results","publication_year":2026,"publication_date":"2026-08-26","ids":{"openalex":"https://openalex.org/W7204603595","doi":"https://doi.org/10.48550/arxiv.2608.26218"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.26218","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.26218","pdf_url":"https://arxiv.org/pdf/2608.26218","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.26218","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129557345","display_name":"Sydney Lewis","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Lewis, Sydney","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5129557345"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":9,"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/T10260","display_name":"Software Engineering Research","score":0.9404000043869019,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10260","display_name":"Software Engineering Research","score":0.9404000043869019,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10533","display_name":"Teaching and Learning Programming","score":0.01080000028014183,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"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/T10743","display_name":"Software Testing and Debugging Techniques","score":0.00930000003427267,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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/ai-coding-agents","display_name":"coding agents","score":0.9769999980926514},{"id":"https://openalex.org/keywords/harness-design","display_name":"harness design","score":0.593999981880188},{"id":"https://openalex.org/keywords/tool-use","display_name":"tool use","score":0.36800000071525574},{"id":"https://openalex.org/keywords/swe-bench","display_name":"SWE-bench","score":0.9049999713897705},{"id":"https://openalex.org/keywords/prompt-engineering","display_name":"prompt engineering","score":0.546999990940094},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.45100000500679016}],"concepts":[{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5213000178337097},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.4964999854564667},{"id":"https://openalex.org/C72563966","wikidata":"https://www.wikidata.org/wiki/Q1303415","display_name":"Cohort","level":2,"score":0.49459999799728394},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.45739999413490295},{"id":"https://openalex.org/C90329073","wikidata":"https://www.wikidata.org/wiki/Q914232","display_name":"Ask price","level":2,"score":0.3935000002384186},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.37310001254081726},{"id":"https://openalex.org/C2777200299","wikidata":"https://www.wikidata.org/wiki/Q52943","display_name":"Conversation","level":2,"score":0.32679998874664307},{"id":"https://openalex.org/C149629883","wikidata":"https://www.wikidata.org/wiki/Q660926","display_name":"Fraction (chemistry)","level":2,"score":0.29019999504089355},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.2892000079154968},{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.26660001277923584},{"id":"https://openalex.org/C102959455","wikidata":"https://www.wikidata.org/wiki/Q7313954","display_name":"Repeated measures design","level":2,"score":0.2653000056743622},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.25949999690055847},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.2590000033378601},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2567000091075897}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.26218","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.26218","pdf_url":"https://arxiv.org/pdf/2608.26218","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.26218","doi":"https://doi.org/10.48550/arxiv.2608.26218","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.26218","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.26218","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.26218","pdf_url":"https://arxiv.org/pdf/2608.26218","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204603595.pdf","grobid_xml":"https://content.openalex.org/works/W7204603595.grobid-xml"},"referenced_works_count":4,"referenced_works":["https://openalex.org/W4387561453","https://openalex.org/W4415252687","https://openalex.org/W7128715130","https://openalex.org/W7139913273"],"related_works":["https://openalex.org/W7171537577","https://openalex.org/W7220707912","https://openalex.org/W7213605503","https://openalex.org/W7213372153","https://openalex.org/W7155774493","https://openalex.org/W7158100772","https://openalex.org/W7214751062","https://openalex.org/W7131850018","https://openalex.org/W7167641297","https://openalex.org/W7216310052"],"abstract_inverted_index":{"A":[0],"coding":[1,51],"agent":[2],"combines":[3],"a":[4,7,119,123,188],"model":[5,13,36,215,224],"with":[6,168],"harness,":[8],"which":[9,15],"decides":[10],"what":[11,213],"the":[12,22,29,32,35,46,56,63,66,76,89,152,161,172,196,210,223,229],"sees,":[14],"tools":[16],"it":[17],"can":[18],"use,":[19],"and":[20,37,79,102,109,122,142,183,225],"how":[21],"work":[23],"continues.":[24],"We":[25,41],"ask":[26],"whether":[27],"changing":[28,209],"harness":[30,48,211,226],"changes":[31],"result":[33],"when":[34],"task":[38],"stay":[39],"fixed.":[40],"compare":[42],"two":[43],"configurations":[44],"of":[45],"same":[47,67,153,162],"on":[49,106,128,160,181],"three":[50,99,165],"benchmarks.":[52],"The":[53,112],"control":[54],"supplies":[55],"full":[57],"conversation":[58],"in":[59,97],"time":[60],"order,":[61],"while":[62,185],"treatment":[64,90,131,155,200],"keeps":[65],"record":[68],"but":[69],"mechanically":[70],"shortens":[71],"older":[72],"tool":[73],"results":[74],"as":[75,228],"context":[77],"fills":[78],"responds":[80],"to":[81,139,147],"repeated":[82],"or":[83],"stalled":[84],"work.":[85],"Under":[86],"tight":[87],"context,":[88],"raises":[91,132,157],"mean":[92,133,190],"per-task":[93,134,191],"fail-to-pass":[94],"fraction":[95],"(F2PF)":[96],"all":[98],"pressure":[100],"comparisons":[101],"increases":[103],"complete":[104,143],"solutions":[105,144],"SWE-bench":[107,110],"Verified":[108,114,182,198],"Pro.":[111],"tight-window":[113],"comparison":[115],"uses":[116],"169":[117],"tasks,":[118],"20,480-token":[120],"window,":[121],"fixed":[124],"480-second":[125],"attempt":[126],"endpoint;":[127],"this":[129],"cohort,":[130,199],"F2PF":[135,192],"from":[136,145],"28":[137],"percent":[138,141],"49":[140],"43":[146],"72.":[148],"Without":[149],"model-specific":[150],"retuning,":[151],"frozen":[154],"also":[156,201],"both":[158],"endpoints":[159],"cohort":[163],"for":[164],"additional":[166],"models":[167],"different":[169],"designs.":[170],"In":[171],"wide-window":[173,197],"Qwen3.6":[174],"comparisons,":[175],"observed":[176],"arm":[177],"outcomes":[178],"are":[179],"close":[180],"Pro,":[184],"FeatureBench":[186],"retains":[187],"higher":[189],"under":[193],"treatment.":[194],"On":[195],"serves":[202],"fewer":[203],"prompt":[204],"tokens":[205],"per":[206],"turn.":[207],"Because":[208],"changed":[212],"unchanged":[214],"weights":[216],"could":[217],"accomplish,":[218],"coding-agent":[219],"evaluations":[220],"should":[221],"treat":[222],"together":[227],"tested":[230],"solver.":[231]},"counts_by_year":[{"year":2026,"cited_by_count":9}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-08-29T00:00:00"}