{"id":"https://openalex.org/W7225137459","doi":"https://doi.org/10.21203/rs.3.rs-11597011/v1","title":"CAPYBALA: Context-Aware Pruning for Yielding and Backtrack-Aligned Loop Avoidance in Autonomous Software Engineering Agents","display_name":"CAPYBALA: Context-Aware Pruning for Yielding and Backtrack-Aligned Loop Avoidance in Autonomous Software Engineering Agents","publication_year":2026,"publication_date":"2026-11-03","ids":{"openalex":"https://openalex.org/W7225137459","doi":"https://doi.org/10.21203/rs.3.rs-11597011/v1"},"language":"en","primary_location":{"id":"doi:10.21203/rs.3.rs-11597011/v1","doi":"https://doi.org/10.21203/rs.3.rs-11597011/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-11597011/v1","pdf_url":"https://www.researchsquare.com/article/rs-11597011/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"},"volume":null,"type":"preprint","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://www.researchsquare.com/article/rs-11597011/latest.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5159167579","display_name":"Khairan Noor Fadhlillah","orcid":"https://orcid.org/0009-0001-8967-3425","observed_orcids":["https://orcid.org/0009-0001-8967-3425"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Khairan Noor Fadhlillah","raw_affiliation_strings":["Universitas Ary Ginanjar"],"raw_orcid":"https://orcid.org/0009-0001-8967-3425","affiliations":[{"raw_affiliation_string":"Universitas Ary Ginanjar","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5159167579"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T10260","display_name":"Software Engineering Research","score":0.763700008392334,"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":0.763700008392334,"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/T10639","display_name":"Advanced Software Engineering Methodologies","score":0.05889999866485596,"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/T10430","display_name":"Software Engineering Techniques and Practices","score":0.04010000079870224,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/token-efficiency","display_name":"token efficiency","score":0.8259999752044678},{"id":"https://openalex.org/keywords/trajectory-dynamics","display_name":"trajectory dynamics","score":0.4620000123977661},{"id":"https://openalex.org/keywords/swe-bench","display_name":"SWE-bench","score":0.5950000286102295}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7883999943733215},{"id":"https://openalex.org/C2779110517","wikidata":"https://www.wikidata.org/wiki/Q1240788","display_name":"Supervisor","level":2,"score":0.5952000021934509},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5134999752044678},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.4767000079154968},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.4738999903202057},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.4611999988555908},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.4390999972820282},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.4284000098705292},{"id":"https://openalex.org/C2780385302","wikidata":"https://www.wikidata.org/wiki/Q367158","display_name":"Protocol (science)","level":3,"score":0.4174000024795532},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3953000009059906},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.37560001015663147},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37040001153945923},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.35899999737739563},{"id":"https://openalex.org/C2779151265","wikidata":"https://www.wikidata.org/wiki/Q1156791","display_name":"Copying","level":2,"score":0.32010000944137573},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.311599999666214},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.29980000853538513},{"id":"https://openalex.org/C5894958","wikidata":"https://www.wikidata.org/wiki/Q2297769","display_name":"Software agent","level":2,"score":0.299699991941452},{"id":"https://openalex.org/C6683253","wikidata":"https://www.wikidata.org/wiki/Q7075535","display_name":"Obstacle avoidance","level":4,"score":0.29899999499320984},{"id":"https://openalex.org/C13687954","wikidata":"https://www.wikidata.org/wiki/Q4826847","display_name":"Autonomous agent","level":2,"score":0.2985999882221222},{"id":"https://openalex.org/C109751979","wikidata":"https://www.wikidata.org/wiki/Q998767","display_name":"Failover","level":2,"score":0.29739999771118164},{"id":"https://openalex.org/C2780704645","wikidata":"https://www.wikidata.org/wiki/Q9251458","display_name":"Observer (physics)","level":2,"score":0.27160000801086426},{"id":"https://openalex.org/C41550386","wikidata":"https://www.wikidata.org/wiki/Q529909","display_name":"Multi-agent system","level":2,"score":0.27149999141693115},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.2676999866962433},{"id":"https://openalex.org/C12269588","wikidata":"https://www.wikidata.org/wiki/Q132364","display_name":"Communications protocol","level":2,"score":0.2639999985694885},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2524000108242035}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21203/rs.3.rs-11597011/v1","doi":"https://doi.org/10.21203/rs.3.rs-11597011/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-11597011/v1","pdf_url":"https://www.researchsquare.com/article/rs-11597011/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"}],"best_oa_location":{"id":"doi:10.21203/rs.3.rs-11597011/v1","doi":"https://doi.org/10.21203/rs.3.rs-11597011/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-11597011/v1","pdf_url":"https://www.researchsquare.com/article/rs-11597011/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7225137459.pdf","grobid_xml":"https://content.openalex.org/works/W7225137459.grobid-xml"},"referenced_works_count":4,"referenced_works":["https://openalex.org/W2626778328","https://openalex.org/W3177813494","https://openalex.org/W4387636003","https://openalex.org/W7103754783"],"related_works":["https://openalex.org/W7212301470","https://openalex.org/W7161745641","https://openalex.org/W7166801336","https://openalex.org/W7167421078","https://openalex.org/W7165691687","https://openalex.org/W7224136098","https://openalex.org/W7224106034","https://openalex.org/W7223553335","https://openalex.org/W7221933441","https://openalex.org/W7221462172"],"abstract_inverted_index":{"Abstract":[0],"Autonomous":[1],"software":[2],"engineering":[3],"agents":[4],"frequently":[5],"become":[6],"entrapped":[7],"in":[8,175],"catastrophic":[9],"error":[10],"loops":[11],"and":[12,51,61,86,112,171,179],"cognitive":[13],"deadlocks,":[14],"burning":[15],"millions":[16],"of":[17,106,155,165],"tokens":[18,157,159],"while":[19,40],"executing":[20],"repetitive":[21],"failing":[22],"tool":[23,49],"invocations.":[24],"Existing":[25],"self-reflection":[26],"frameworks":[27],"exacerbate":[28],"token":[29],"exhaustion":[30],"by":[31],"appending":[32],"redundant":[33],"verbal":[34],"history":[35],"to":[36,138,142,147,173],"the":[37,117],"context":[38,42,89],"window,":[39],"naive":[41],"truncation":[43],"routinely":[44],"violates":[45],"protocol":[46,93],"invariants":[47],"between":[48],"calls":[50],"execution":[52,84,114],"responses.":[53],"We":[54],"introduce":[55],"CAPYBALA":[56,133],"(Context-Aware":[57],"Pruning":[58],"for":[59],"Yielding":[60],"Backtrack-Aligned":[62],"Loop":[63],"Avoidance),":[64],"an":[65,101,152],"in-memory":[66],"runtime":[67],"supervisor":[68,189],"that":[69],"tracks":[70],"agent":[71,110],"trajectory":[72,103,121],"dynamics":[73],"via":[74],"a":[75,181],"signed":[76],"feature-hashed":[77],"phase-space":[78],"manifold":[79],"(O(1)":[80],"space":[81],"complexity,":[82],"sub-millisecond":[83,182],"latency)":[85],"performs":[87],"atomic":[88],"pruning":[90],"with":[91,136],"formal":[92],"pairing":[94],"guarantees":[95],"(N_tool_call":[96],"=":[97,192,201,205],"N_tool_response).":[98],"Evaluated":[99],"across":[100,160,168],"offline":[102],"replay":[104],"benchmark":[105],"200":[107,161],"real-world":[108],"SWE-bench":[109],"runs":[111],"3,209":[113],"turns":[115,167],"from":[116],"open":[118],"Nebius":[119],"SWE-agent":[120],"corpus,":[122],"alongside":[123],"live":[124,176],"head-to-head":[125],"coding":[126,177],"duels":[127],"on":[128],"active":[129],"commercial":[130],"foundation":[131],"models,":[132],"captures":[134],"deadlocks":[135],"86.00%":[137],"97.86%":[139],"recall,":[140],"95.56%":[141],"98.56%":[143],"precision":[144],"(with":[145],"96.00%":[146],"96.67%":[148],"clean":[149],"specificity),":[150],"eliminates":[151],"aggregate":[153],"42.8%":[154],"wasted":[156],"(1,017,250":[158],"trajectories,":[162],"saving":[163],"59.74%":[164],"non-productive":[166],"stuck":[169],"trajectories":[170],"40.3%":[172],"47.1%":[174],"duels),":[178],"maintains":[180],"mean":[183,188],"decision":[184],"latency":[185],"(0.641":[186],"ms":[187,207],"turn,":[190],"p50":[191,200],"0.546":[193],"ms;":[194],"53":[195],"\u03bcs":[196],"raw":[197],"feature":[198],"extraction,":[199],"38":[202],"\u03bcs;":[203],"p99":[204],"5.84":[206],"tail)":[208],"without":[209],"requiring":[210],"model":[211],"fine-tuning":[212],"or":[213],"secondary":[214],"embedding":[215],"inference.":[216]},"counts_by_year":[],"updated_date":"2026-11-05T12:25:35.769452","created_date":"2026-11-04T00:00:00"}