{"id":"https://openalex.org/W7224772147","doi":"https://doi.org/10.48550/arxiv.2610.38566","title":"Co-Evolving Policy and Rubric via Capability Awareness for Open-Ended Generation","display_name":"Co-Evolving Policy and Rubric via Capability Awareness for Open-Ended Generation","publication_year":2026,"publication_date":"2026-10-28","ids":{"openalex":"https://openalex.org/W7224772147","doi":"https://doi.org/10.48550/arxiv.2610.38566"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.38566","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.38566","pdf_url":"https://arxiv.org/pdf/2610.38566","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/2610.38566","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5160236688","display_name":"Chengsheng Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]},{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhang, Chengsheng","raw_affiliation_strings":["Alibaba Token Hub , Alibaba Group","University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Token Hub , Alibaba Group","institution_ids":["https://openalex.org/I45928872"]},{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100299294","display_name":"Wenshen Xu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu, Wenshen","raw_affiliation_strings":["Alibaba Token Hub , Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Token Hub , Alibaba Group","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160217870","display_name":"Yuchuan Wu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wu, Yuchuan","raw_affiliation_strings":["Alibaba Token Hub , Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Token Hub , Alibaba Group","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160196423","display_name":"Zhiyuan Yu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]},{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu, Zhiyuan","raw_affiliation_strings":["Alibaba Token Hub , Alibaba Group","Nanjing University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Token Hub , Alibaba Group","institution_ids":["https://openalex.org/I45928872"]},{"raw_affiliation_string":"Nanjing University","institution_ids":["https://openalex.org/I881766915"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160215920","display_name":"Junfeng Fang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Fang, Junfeng","raw_affiliation_strings":["National University of Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160211514","display_name":"Tianlong Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Tianlong","raw_affiliation_strings":["Alibaba Token Hub , Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Token Hub , Alibaba Group","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5104202404","display_name":"Bingxue Qiu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qiu, Bingxue","raw_affiliation_strings":["Alibaba Token Hub , Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Token Hub , Alibaba Group","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160168195","display_name":"Yongbin Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li, Yongbin","raw_affiliation_strings":["Alibaba Token Hub , Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Token Hub , Alibaba Group","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5160233525","display_name":"Xiang Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Xiang","raw_affiliation_strings":["University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.4562999904155731,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.4562999904155731,"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":0.21240000426769257,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.13459999859333038,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.7089999914169312},{"id":"https://openalex.org/keywords/open-ended-generation","display_name":"open-ended generation","score":0.9160000085830688},{"id":"https://openalex.org/keywords/self-distillation","display_name":"self-distillation","score":0.8040000200271606},{"id":"https://openalex.org/keywords/capability-gaps","display_name":"capability gaps","score":0.5339999794960022}],"concepts":[{"id":"https://openalex.org/C111640148","wikidata":"https://www.wikidata.org/wiki/Q847349","display_name":"Rubric","level":2,"score":0.9505000114440918},{"id":"https://openalex.org/C2781313679","wikidata":"https://www.wikidata.org/wiki/Q66793593","display_name":"Coca","level":2,"score":0.6460000276565552},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6274999976158142},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.4632999897003174},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.3871000111103058},{"id":"https://openalex.org/C205711294","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Rendering (computer graphics)","level":2,"score":0.3652999997138977},{"id":"https://openalex.org/C195094911","wikidata":"https://www.wikidata.org/wiki/Q14167904","display_name":"Process management","level":1,"score":0.33079999685287476},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.32690000534057617},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.29170000553131104},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.28290000557899475},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.271699994802475}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.38566","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.38566","pdf_url":"https://arxiv.org/pdf/2610.38566","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.2610.38566","doi":"https://doi.org/10.48550/arxiv.2610.38566","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.38566","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:2610.38566","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.38566","pdf_url":"https://arxiv.org/pdf/2610.38566","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/W7224772147.pdf","grobid_xml":"https://content.openalex.org/works/W7224772147.grobid-xml"},"referenced_works_count":31,"referenced_works":["https://openalex.org/W4391631327","https://openalex.org/W4403883066","https://openalex.org/W4411630288","https://openalex.org/W4414281281","https://openalex.org/W4414490998","https://openalex.org/W4414939355","https://openalex.org/W4415109214","https://openalex.org/W4416035050","https://openalex.org/W4417006854","https://openalex.org/W6891853064","https://openalex.org/W6929370958","https://openalex.org/W6947991961","https://openalex.org/W7117549943","https://openalex.org/W7127380112","https://openalex.org/W7127948103","https://openalex.org/W7130402983","https://openalex.org/W7140517510","https://openalex.org/W7150828934","https://openalex.org/W7160378838","https://openalex.org/W7160838788","https://openalex.org/W7160917236","https://openalex.org/W7160918678","https://openalex.org/W7161046900","https://openalex.org/W7162788548","https://openalex.org/W7164002160","https://openalex.org/W7164539611","https://openalex.org/W7165692494","https://openalex.org/W7167570954","https://openalex.org/W7169497493","https://openalex.org/W7169884067","https://openalex.org/W7170175357"],"related_works":["https://openalex.org/W7165692494","https://openalex.org/W7162788548","https://openalex.org/W7206162831","https://openalex.org/W7221779024","https://openalex.org/W7223038803","https://openalex.org/W7202174126","https://openalex.org/W7170175357","https://openalex.org/W7220679763","https://openalex.org/W7224869484","https://openalex.org/W7223027090"],"abstract_inverted_index":{"Rubric-based":[0],"reinforcement":[1],"learning":[2],"(RL)":[3],"offers":[4],"interpretable,":[5],"fine-grained":[6],"supervision":[7],"for":[8,130],"aligning":[9],"large":[10],"language":[11],"models":[12],"(LLMs)":[13],"on":[14,134],"open-ended":[15],"generation":[16],"tasks.":[17],"However,":[18],"static":[19],"rubric":[20,55,98],"criteria":[21,121],"may":[22],"become":[23],"saturated":[24],"or":[25,45,145],"remain":[26],"beyond":[27],"the":[28,32,59,69,73,97,117,153,158,168,191],"policy's":[29,61,74],"exploration,":[30],"rendering":[31],"reward":[33],"signals":[34],"insufficiently":[35],"discriminative.":[36],"While":[37],"recent":[38],"efforts":[39],"seek":[40],"to":[41,50,111,116],"guide":[42],"sampling":[43],"exploration":[44],"evolve":[46],"rubrics,":[47],"they":[48],"fail":[49],"explicitly":[51],"factor":[52],"in":[53,64],"how":[54],"difficulty":[56],"should":[57],"track":[58],"evolving":[60],"capabilities,":[62],"resulting":[63],"a":[65,91],"persistent":[66],"mismatch":[67],"between":[68],"optimization":[70,144],"objective":[71],"and":[72,85,99,108,125,161,176],"current":[75,118],"capability":[76,128,159,201],"frontier.":[77],"In":[78],"this":[79,135],"paper,":[80],"we":[81],"propose":[82],"\\textbf{Co}-Evolving":[83],"Policy":[84],"Rubric":[86],"via":[87],"\\textbf{C}apability":[88],"\\textbf{A}wareness":[89],"(\\textbf{CoCA}),":[90],"closed-loop":[92],"framework":[93],"that":[94,179],"jointly":[95],"adapts":[96],"policy":[100,154],"through":[101,142,149],"criterion-level":[102],"diagnosis.":[103],"Specifically,":[104],"CoCA":[105,137,156,180,189],"contrasts":[106],"query-only":[107,197],"rubric-guided":[109],"rollouts":[110],"characterize":[112],"each":[113],"criterion":[114],"relative":[115],"policy,":[119],"recalibrates":[120],"with":[122],"mismatched":[123],"difficulty,":[124],"identifies":[126],"guidance-revealed":[127,194],"gaps":[129],"frontier-based":[131],"learning.":[132],"Based":[133],"insight,":[136],"either":[138],"expands":[139],"rubric-conditioned":[140],"frontiers":[141],"guided":[143],"transfers":[146],"qualified":[147],"gains":[148],"on-policy":[150],"self-distillation.":[151],"As":[152],"updates,":[155],"improves":[157,182],"diagnosis":[160],"enables":[162],"more":[163],"reliable":[164],"distillation,":[165],"thereby":[166],"completing":[167],"co-evolution":[169],"loop.":[170],"Extensive":[171],"experiments":[172],"across":[173],"diverse":[174],"benchmarks":[175],"LLMs":[177],"demonstrate":[178],"consistently":[181],"performance":[183],"over":[184],"rubric-based":[185],"RL":[186],"methods.":[187],"Moreover,":[188],"promotes":[190],"internalization":[192],"of":[193],"capabilities":[195],"into":[196],"behavior":[198],"while":[199],"mitigating":[200],"degradation.":[202]},"counts_by_year":[],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2026-10-31T00:00:00"}