{"id":"https://openalex.org/W4416246346","doi":"https://doi.org/10.48550/arxiv.2511.10507","title":"AdvancedIF: Rubric-Based Benchmarking and Reinforcement Learning for Advancing LLM Instruction Following","display_name":"AdvancedIF: Rubric-Based Benchmarking and Reinforcement Learning for Advancing LLM Instruction Following","publication_year":2025,"publication_date":"2025-11-13","ids":{"openalex":"https://openalex.org/W4416246346","doi":"https://doi.org/10.48550/arxiv.2511.10507"},"language":null,"primary_location":{"id":"pmh:oai:arXiv.org:2511.10507","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2511.10507","pdf_url":"https://arxiv.org/pdf/2511.10507","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2511.10507","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101467216","display_name":"Yun He","orcid":"https://orcid.org/0000-0001-9573-7678","observed_orcids":["https://orcid.org/0000-0001-9573-7678"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Yun","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100691367","display_name":"Wenzhe Li","orcid":"https://orcid.org/0000-0002-0153-7999","observed_orcids":["https://orcid.org/0000-0002-0153-7999"]},"institutions":[{"id":"https://openalex.org/I20089843","display_name":"Princeton University","ror":"https://ror.org/00hx57361","country_code":"US","type":"education","lineage":["https://openalex.org/I20089843"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Li, Wenzhe","raw_affiliation_strings":["Princeton University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Princeton University ,","institution_ids":["https://openalex.org/I20089843"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047725424","display_name":"Hejia Zhang","orcid":"https://orcid.org/0000-0001-5222-5277","observed_orcids":["https://orcid.org/0000-0001-5222-5277"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Hejia","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100376091","display_name":"Song\u2010Lin Li","orcid":"https://orcid.org/0000-0003-1354-7729","observed_orcids":["https://orcid.org/0000-0003-1354-7729"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Songlin","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078338025","display_name":"Karishma Mandyam","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mandyam, Karishma","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025458684","display_name":"Sopan Khosla","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Khosla, Sopan","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056336934","display_name":"Yuanhao Xiong","orcid":"https://orcid.org/0000-0002-0940-2079","observed_orcids":["https://orcid.org/0000-0002-0940-2079"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiong, Yuanhao","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039996365","display_name":"Nianyi Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Nanshu","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Peng, Xiaoliang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng, Xiaoliang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028849793","display_name":"Beibin Li","orcid":"https://orcid.org/0000-0002-4426-3449","observed_orcids":["https://orcid.org/0000-0002-4426-3449"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Beibin","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Bi, Shengjie","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bi, Shengjie","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022259826","display_name":"Shishir G. Patil","orcid":"https://orcid.org/0000-0001-6835-388X","observed_orcids":["https://orcid.org/0000-0001-6835-388X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Patil, Shishir G.","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Qi, Qi","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qi, Qi","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056967366","display_name":"Shengyu Feng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Feng, Shengyu","raw_affiliation_strings":["CMU"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CMU","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066883909","display_name":"Julian Katz-Samuels","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Katz-Samuels, Julian","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103936332","display_name":"Richard Yuanzhe Pang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pang, Richard Yuanzhe","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030673462","display_name":"Sujan K. Gonugondla","orcid":"https://orcid.org/0000-0003-4743-6461","observed_orcids":["https://orcid.org/0000-0003-4743-6461"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gonugondla, Sujan","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041547937","display_name":"Hunter Lang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lang, Hunter","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011105950","display_name":"Yue Yu","orcid":"https://orcid.org/0000-0003-3452-8391","observed_orcids":["https://orcid.org/0000-0003-3452-8391"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Yue","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003241968","display_name":"Yundi Qian","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qian, Yundi","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053471950","display_name":"Maryam Fazel-Zarandi","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fazel-Zarandi, Maryam","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036418431","display_name":"Licheng Yu","orcid":"https://orcid.org/0000-0002-4943-6732","observed_orcids":["https://orcid.org/0000-0002-4943-6732"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Licheng","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063231528","display_name":"Amine Benhalloum","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Benhalloum, Amine","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030937723","display_name":"Hany Hassan Awadalla","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Awadalla, Hany","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5014855689","display_name":"Manaal Faruqui","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Faruqui, Manaal","raw_affiliation_strings":["Meta Superintelligence Labs ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meta Superintelligence Labs ,","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"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":0.6176999807357788,"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":0.6176999807357788,"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":0.21619999408721924,"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/T12031","display_name":"Speech and dialogue systems","score":0.04610000178217888,"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/instruction-following","display_name":"instruction following","score":0.7829999923706055},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.9490000009536743},{"id":"https://openalex.org/keywords/rubrics","display_name":"rubrics","score":0.6539999842643738},{"id":"https://openalex.org/keywords/benchmarking","display_name":"benchmarking","score":0.8429999947547913},{"id":"https://openalex.org/keywords/system-prompts","display_name":"system prompts","score":0.8429999947547913},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"reinforcement learning","score":0.5699999928474426},{"id":"https://openalex.org/keywords/reward-shaping","display_name":"reward shaping","score":0.9819999933242798},{"id":"https://openalex.org/keywords/post-training-fine-tuning","display_name":"post-training fine-tuning","score":0.8349999785423279},{"id":"https://openalex.org/keywords/llm-evaluation","display_name":"LLM evaluation","score":0.6990000009536743}],"concepts":[{"id":"https://openalex.org/C111640148","wikidata":"https://www.wikidata.org/wiki/Q847349","display_name":"Rubric","level":2,"score":0.9314000010490417},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.8181999921798706},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7617999911308289},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6974999904632568},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6525999903678894},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.6438000202178955},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.4447999894618988},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4041999876499176},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.3774000108242035},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.34279999136924744},{"id":"https://openalex.org/C2776904630","wikidata":"https://www.wikidata.org/wiki/Q356336","display_name":"Adept","level":3,"score":0.33090001344680786},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.31690001487731934},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.2897999882698059},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.27639999985694885},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.27239999175071716},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2583000063896179},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.25369998812675476}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2511.10507","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2511.10507","pdf_url":"https://arxiv.org/pdf/2511.10507","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2511.10507","doi":"https://doi.org/10.48550/arxiv.2511.10507","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2511.10507","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2511.10507","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2511.10507","pdf_url":"https://arxiv.org/pdf/2511.10507","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416246346.pdf","grobid_xml":"https://content.openalex.org/works/W4416246346.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"progress":[1],"in":[2,150,167],"large":[3],"language":[4],"models":[5],"(LLMs)":[6],"has":[7],"led":[8],"to":[9,79,108],"impressive":[10],"performance":[11],"on":[12,133,138],"a":[13,28,65,93,101,129,157],"range":[14],"of":[15,44,126,147],"tasks,":[16],"yet":[17],"advanced":[18,165],"instruction":[19,114],"following":[20],"(IF)-especially":[21],"for":[22,36,113,160,172],"complex,":[23,81],"multi-turn,":[24,82],"and":[25,33,48,72,83,105,135,163,175],"system-prompted":[26],"instructions-remains":[27],"significant":[29],"challenge.":[30],"Rigorous":[31],"evaluation":[32],"effective":[34,110],"training":[35,162],"such":[37],"capabilities":[38],"are":[39],"hindered":[40],"by":[41],"the":[42,123,145,170],"lack":[43],"high-quality,":[45],"human-annotated":[46],"benchmarks":[47],"reliable,":[49],"interpretable":[50],"reward":[51,106],"signals.":[52],"In":[53],"this":[54,62],"work,":[55],"we":[56],"introduce":[57],"AdvancedIF":[58,134],"(we":[59],"will":[60],"release":[61],"benchmark":[63,67],"soon),":[64],"comprehensive":[66],"featuring":[68],"over":[69],"1,600":[70],"prompts":[71],"expert-curated":[73],"rubrics":[74,155],"that":[75,97,119],"assess":[76],"LLMs":[77],"ability":[78],"follow":[80],"system-level":[84],"instructions.":[85],"We":[86],"further":[87],"propose":[88],"RIFL":[89,120],"(Rubric-based":[90],"Instruction-Following":[91],"Learning),":[92],"novel":[94],"post-training":[95],"pipeline":[96],"leverages":[98],"rubric":[99,103],"generation,":[100],"finetuned":[102],"verifier,":[104],"shaping":[107],"enable":[109],"reinforcement":[111],"learning":[112],"following.":[115],"Extensive":[116],"experiments":[117],"demonstrate":[118],"substantially":[121],"improves":[122],"instruction-following":[124],"abilities":[125],"LLMs,":[127,168],"achieving":[128],"6.7%":[130],"absolute":[131],"gain":[132],"strong":[136],"results":[137],"public":[139],"benchmarks.":[140],"Our":[141],"ablation":[142],"studies":[143],"confirm":[144],"effectiveness":[146],"each":[148],"component":[149],"RIFL.":[151],"This":[152],"work":[153],"establishes":[154],"as":[156],"powerful":[158],"tool":[159],"both":[161],"evaluating":[164],"IF":[166],"paving":[169],"way":[171],"more":[173],"capable":[174],"reliable":[176],"AI":[177],"systems.":[178]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-10-12T09:40:04.860316","created_date":"2025-11-15T00:00:00"}