{"id":"https://openalex.org/W4412945019","doi":"https://doi.org/10.18653/v1/2025.acl-long.880","title":"ICR Probe: Tracking Hidden State Dynamics for Reliable Hallucination Detection in LLMs","display_name":"ICR Probe: Tracking Hidden State Dynamics for Reliable Hallucination Detection in LLMs","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412945019","doi":"https://doi.org/10.18653/v1/2025.acl-long.880"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.880","doi":"https://doi.org/10.18653/v1/2025.acl-long.880","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.880","pdf_url":"https://aclanthology.org/2025.acl-long.880.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://aclanthology.org/2025.acl-long.880.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5107875723","display_name":"Zhenliang Zhang","orcid":"https://orcid.org/0009-0002-1448-8505","observed_orcids":["https://orcid.org/0009-0002-1448-8505"]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhenliang Zhang","raw_affiliation_strings":["School of Software and Microelectronics , Peking University","Wangxuan Institute of Computer Technology , Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Software and Microelectronics , Peking University","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"Wangxuan Institute of Computer Technology , Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069542066","display_name":"Xinyu Hu","orcid":"https://orcid.org/0000-0002-9739-5613","observed_orcids":["https://orcid.org/0000-0002-9739-5613","https://orcid.org/0000-0001-7463-7467"]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xinyu Hu","raw_affiliation_strings":["Wangxuan Institute of Computer Technology , Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wangxuan Institute of Computer Technology , Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Huixuan Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huixuan Zhang","raw_affiliation_strings":["Wangxuan Institute of Computer Technology , Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wangxuan Institute of Computer Technology , Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101956300","display_name":"Junzhe Zhang","orcid":"https://orcid.org/0000-0003-1931-8046","observed_orcids":["https://orcid.org/0000-0003-1931-8046"]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Junzhe Zhang","raw_affiliation_strings":["Wangxuan Institute of Computer Technology , Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wangxuan Institute of Computer Technology , Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5029568096","display_name":"Xiaojun Wan","orcid":"https://orcid.org/0000-0001-6887-1994","observed_orcids":["https://orcid.org/0000-0001-6887-1994"]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaojun Wan","raw_affiliation_strings":["Wangxuan Institute of Computer Technology , Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wangxuan Institute of Computer Technology , Peking University","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I20231570"],"apc_list":null,"apc_paid":null,"fwci":5.649,"has_fulltext":true,"cited_by_count":6,"citation_normalized_percentile":{"value":0.96517539,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":90,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"17986","last_page":"18002"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.6845999956130981,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.6845999956130981,"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":0.12290000170469284,"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":0.012799999676644802,"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/hallucination-detection","display_name":"hallucination detection","score":0.9729999899864197},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.9350000023841858},{"id":"https://openalex.org/keywords/hidden-state-dynamics","display_name":"hidden state dynamics","score":0.9679999947547913}],"concepts":[{"id":"https://openalex.org/C2775936607","wikidata":"https://www.wikidata.org/wiki/Q466845","display_name":"Tracking (education)","level":2,"score":0.6340132355690002},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.49173954129219055},{"id":"https://openalex.org/C145912823","wikidata":"https://www.wikidata.org/wiki/Q113558","display_name":"Dynamics (music)","level":2,"score":0.48882243037223816},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.46691277623176575},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42808467149734497},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.1760905385017395},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.08585959672927856},{"id":"https://openalex.org/C19417346","wikidata":"https://www.wikidata.org/wiki/Q7922","display_name":"Pedagogy","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.880","doi":"https://doi.org/10.18653/v1/2025.acl-long.880","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.880","pdf_url":"https://aclanthology.org/2025.acl-long.880.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2507.16488","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2507.16488","pdf_url":"https://arxiv.org/pdf/2507.16488","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.2507.16488","doi":"https://doi.org/10.48550/arxiv.2507.16488","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2507.16488","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/v1/2025.acl-long.880","doi":"https://doi.org/10.18653/v1/2025.acl-long.880","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.880","pdf_url":"https://aclanthology.org/2025.acl-long.880.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412945019.pdf","grobid_xml":"https://content.openalex.org/works/W4412945019.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7167724835","https://openalex.org/W4415088582","https://openalex.org/W4415317673","https://openalex.org/W4414360390","https://openalex.org/W7155060417","https://openalex.org/W7155032038","https://openalex.org/W4408352939","https://openalex.org/W7172526003","https://openalex.org/W4417491397","https://openalex.org/W7215938166"],"abstract_inverted_index":{"Large":[0],"language":[1,8],"models":[2],"(LLMs)":[3],"excel":[4],"at":[5],"various":[6],"natural":[7],"processing":[9],"tasks,":[10],"but":[11],"their":[12,18,34],"tendency":[13],"to":[14,49,65,74],"generate":[15],"hallucinations":[16],"undermines":[17],"reliability.Existing":[19],"hallucination":[20,98],"detection":[21,99],"methods":[22],"leveraging":[23],"hidden":[24,51,76,110],"states":[25],"predominantly":[26],"focus":[27,48],"on":[28,92],"static":[29],"and":[30,55,87,127],"isolated":[31],"representations,":[32],"overlooking":[33],"dynamic":[35],"evolution":[36,108],"across":[37],"layers,":[38],"which":[39,68,104],"limits":[40],"efficacy.To":[41],"address":[42],"this":[43,138],"limitation,":[44],"we":[45,95],"shift":[46],"the":[47,50,60,70,75,82,101,106,115,134],"state":[52],"update":[53],"process":[54],"introduce":[56],"a":[57,97],"novel":[58],"metric,":[59],"ICR":[61,83,102,116],"Score":[62,84],"(Information":[63],"Contribution":[64],"Residual":[66],"Stream),":[67],"quantifies":[69],"contribution":[71],"of":[72,109,137],"modules":[73],"states'":[77],"update.We":[78],"empirically":[79],"validate":[80],"that":[81,114],"is":[85],"effective":[86],"reliable":[88],"in":[89],"distinguishing":[90],"hallucinations.Building":[91],"these":[93],"insights,":[94],"propose":[96],"method,":[100,139],"Probe,":[103],"captures":[105],"cross-layer":[107],"states.Experimental":[111],"results":[112],"show":[113],"Probe":[117],"achieves":[118],"superior":[119],"performance":[120],"with":[121],"significantly":[122],"fewer":[123],"parameters.Furthermore,":[124],"ablation":[125],"studies":[126],"case":[128],"analyses":[129],"offer":[130],"deeper":[131],"insights":[132],"into":[133],"underlying":[135],"mechanism":[136],"improving":[140],"its":[141],"interpretability.":[142]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":1}],"updated_date":"2026-11-01T11:40:01.658920","created_date":"2025-10-10T00:00:00"}
