{"id":"https://openalex.org/W7204937544","doi":"https://doi.org/10.18653/2026.emnlp-main.522","title":"Evaluating and Improving LLM Self-Modeling","display_name":"Evaluating and Improving LLM Self-Modeling","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7204937544","doi":"https://doi.org/10.18653/2026.emnlp-main.522"},"language":"en","primary_location":{"id":"doi:10.18653/2026.emnlp-main.522","doi":"https://doi.org/10.18653/2026.emnlp-main.522","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.522","pdf_url":"https://aclanthology.org/2026.emnlp-main.522.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","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":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2026.emnlp-main.522.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5017018703","display_name":"Siqi Zeng","orcid":"https://orcid.org/0009-0008-2042-0754","observed_orcids":["https://orcid.org/0009-0008-2042-0754"]},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Siqi Zeng","raw_affiliation_strings":["University of Illinois Urbana-Champaign"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Illinois Urbana-Champaign","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003903937","display_name":"Andre N. Assis","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Andre N. Assis","raw_affiliation_strings":["Constellation"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Constellation","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5148906719","display_name":"Rowan Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rowan Wang","raw_affiliation_strings":["Constellation"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Constellation","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.69751011,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11978","last_page":"12062"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.6211000084877014,"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":0.6211000084877014,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.13050000369548798,"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":0.05990000069141388,"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/self-modeling","display_name":"self-modeling","score":0.8989999890327454},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.9639999866485596},{"id":"https://openalex.org/keywords/self-reflection","display_name":"self-reflection","score":0.9610000252723694},{"id":"https://openalex.org/keywords/counterfactual-thinking","display_name":"counterfactual reasoning","score":0.6610000133514404},{"id":"https://openalex.org/keywords/synthetic-training-data","display_name":"synthetic training data","score":0.7450000047683716},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"reinforcement learning","score":0.8519999980926514}],"concepts":[{"id":"https://openalex.org/C108650721","wikidata":"https://www.wikidata.org/wiki/Q1783253","display_name":"Counterfactual thinking","level":2,"score":0.7875000238418579},{"id":"https://openalex.org/C129671850","wikidata":"https://www.wikidata.org/wiki/Q210501","display_name":"Introspection","level":2,"score":0.666700005531311},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6489999890327454},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5471000075340271},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.4681999981403351},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.45500001311302185},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.45190000534057617},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.4481000006198883},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4408000111579895},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43560001254081726},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.41530001163482666},{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.3637000024318695},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.3506999909877777},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.3343999981880188},{"id":"https://openalex.org/C71889745","wikidata":"https://www.wikidata.org/wiki/Q1783264","display_name":"Counterfactual conditional","level":3,"score":0.32989999651908875},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.30090001225471497},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.29600000381469727},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.2563999891281128}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/2026.emnlp-main.522","doi":"https://doi.org/10.18653/2026.emnlp-main.522","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.522","pdf_url":"https://aclanthology.org/2026.emnlp-main.522.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","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":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2608.30980","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.30980","pdf_url":"https://arxiv.org/pdf/2608.30980","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.2608.30980","doi":"https://doi.org/10.48550/arxiv.2608.30980","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.30980","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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.18653/2026.emnlp-main.522","doi":"https://doi.org/10.18653/2026.emnlp-main.522","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.522","pdf_url":"https://aclanthology.org/2026.emnlp-main.522.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","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":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204937544.pdf","grobid_xml":"https://content.openalex.org/works/W7204937544.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7127332743","https://openalex.org/W4403580095","https://openalex.org/W4399554107","https://openalex.org/W4405094468","https://openalex.org/W7202276272","https://openalex.org/W7162531820","https://openalex.org/W4414486237","https://openalex.org/W7119538405","https://openalex.org/W4406733901","https://openalex.org/W6929297702"],"abstract_inverted_index":{"We":[0],"study":[1],"self-modeling:":[2],"an":[3],"LLM's":[4],"ability":[5],"to":[6,94,102,114],"answer":[7],"questions":[8,58],"about":[9,59],"its":[10],"own":[11,61],"behavior.We":[12],"focus":[13],"on":[14,55,126],"verifiable":[15],"behavioral":[16],"questions,":[17],"such":[18],"as":[19],"whether":[20],"a":[21,35,68],"prompt":[22],"edit":[23],"would":[24],"change":[25],"the":[26,115],"model's":[27,116],"final":[28],"answer.To":[29],"measure":[30],"this":[31],"capability,":[32],"we":[33,66],"introduce":[34],"benchmark":[36],"that":[37,72,79],"tests":[38],"diverse":[39],"types":[40],"of":[41],"self-modeling":[42,64,74,84,107],"questions.Current":[43],"models":[44],"show":[45,78],"non-trivial":[46],"but":[47],"limited":[48],"selfmodeling":[49],"skill,":[50,65],"and":[51,77,123],"make":[52],"systematic":[53],"mistakes":[54],"simple":[56],"counterfactual":[57],"their":[60],"behavior.To":[62],"improve":[63,82],"develop":[67],"scalable":[69],"synthetic-data":[70],"pipeline":[71],"produces":[73],"training":[75,124],"data,":[76],"reinforcement-learning":[80],"can":[81],"aggregate":[83],"skill":[85],"across":[86],"three":[87],"opensource":[88],"model":[89],"families":[90],"with":[91],"some":[92],"transfer":[93],"held-out":[95],"tasks.These":[96],"gains,":[97],"however,":[98],"do":[99],"not":[100,109],"seem":[101],"constitute":[103],"introspection":[104],"consistently:":[105],"improved":[106],"may":[108],"arise":[110],"from":[111],"privileged":[112],"access":[113],"internal":[117],"decision":[118],"process.We":[119],"release":[120],"our":[121],"evaluation":[122],"code":[125],"GitHub.":[127],"1":[128]},"counts_by_year":[],"updated_date":"2026-10-31T12:09:11.096373","created_date":"2026-09-02T00:00:00"}