{"id":"https://openalex.org/W7165056066","doi":"https://doi.org/10.1145/3805713.3820397","title":"RoTRAG: Rule of Thumb Reasoning for Conversation Harm Detection with Retrieval-Augmented Generation","display_name":"RoTRAG: Rule of Thumb Reasoning for Conversation Harm Detection with Retrieval-Augmented Generation","publication_year":2026,"publication_date":"2026-07-14","ids":{"openalex":"https://openalex.org/W7165056066","doi":"https://doi.org/10.1145/3805713.3820397"},"language":null,"primary_location":{"id":"doi:10.1145/3805713.3820397","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805713.3820397","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 International ACM SIGIR Conference on Innovative Concepts and Theories in Information Retrieval (ICTIR)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3805713.3820397","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138815496","display_name":"Juhyeon Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I4210161752","display_name":"Beijing Haidian Hospital","ror":"https://ror.org/058x5eq06","country_code":"CN","type":"healthcare","lineage":["https://openalex.org/I4210161752"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Juhyeon Lee","raw_affiliation_strings":["Peking University, Haidian, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0006-5696-3940","affiliations":[{"raw_affiliation_string":"Peking University, Haidian, Beijing, China","institution_ids":["https://openalex.org/I4210161752"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138812796","display_name":"Wonduk Seo","orcid":null},"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":"Wonduk Seo","raw_affiliation_strings":["Enhans, Seoul, Republic of Korea and Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0008-6070-1833","affiliations":[{"raw_affiliation_string":"Enhans, Seoul, Republic of Korea and Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120278526","display_name":"Junseo Koh","orcid":null},"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":"Junseo Koh","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0008-5965-2130","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091873011","display_name":"S Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I4210091586","display_name":"Yuhan (South Korea)","ror":"https://ror.org/00bv1jr75","country_code":"KR","type":"company","lineage":["https://openalex.org/I4210091586"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seunghyun Lee","raw_affiliation_strings":["Enhans, Seoul, Republic of Korea"],"raw_orcid":"https://orcid.org/0009-0000-1687-1597","affiliations":[{"raw_affiliation_string":"Enhans, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I4210091586"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138775279","display_name":"Haihua Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I123534392","display_name":"University of North Texas","ror":"https://ror.org/00v97ad02","country_code":"US","type":"education","lineage":["https://openalex.org/I123534392"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Haihua Chen","raw_affiliation_strings":["University of North Texas, Denton, TX, USA"],"raw_orcid":"https://orcid.org/0000-0002-7088-9752","affiliations":[{"raw_affiliation_string":"University of North Texas, Denton, TX, USA","institution_ids":["https://openalex.org/I123534392"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5138822100","display_name":"Yi Bu","orcid":null},"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":"Yi Bu","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-2549-4580","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":3,"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":"218","last_page":"229"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.36340001225471497,"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":0.36340001225471497,"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/T12262","display_name":"Hate Speech and Cyberbullying Detection","score":0.21119999885559082,"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":0.06019999831914902,"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/rule-of-thumb","display_name":"Rule of thumb","score":0.6938999891281128},{"id":"https://openalex.org/keywords/conversation","display_name":"Conversation","score":0.5504000186920166},{"id":"https://openalex.org/keywords/harm","display_name":"Harm","score":0.5335000157356262},{"id":"https://openalex.org/keywords/normative","display_name":"Normative","score":0.5008999705314636},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.3538999855518341},{"id":"https://openalex.org/keywords/classifier","display_name":"Classifier (UML)","score":0.3294999897480011},{"id":"https://openalex.org/keywords/binary-classification","display_name":"Binary classification","score":0.31869998574256897}],"concepts":[{"id":"https://openalex.org/C89246107","wikidata":"https://www.wikidata.org/wiki/Q1398821","display_name":"Rule of thumb","level":2,"score":0.6938999891281128},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6471999883651733},{"id":"https://openalex.org/C2777200299","wikidata":"https://www.wikidata.org/wiki/Q52943","display_name":"Conversation","level":2,"score":0.5504000186920166},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5353000164031982},{"id":"https://openalex.org/C2777363581","wikidata":"https://www.wikidata.org/wiki/Q15098235","display_name":"Harm","level":2,"score":0.5335000157356262},{"id":"https://openalex.org/C44725695","wikidata":"https://www.wikidata.org/wiki/Q288156","display_name":"Normative","level":2,"score":0.5008999705314636},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.374099999666214},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.3538999855518341},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.3294999897480011},{"id":"https://openalex.org/C66905080","wikidata":"https://www.wikidata.org/wiki/Q17005494","display_name":"Binary classification","level":3,"score":0.31869998574256897},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.3167000114917755},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.3142000138759613},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3077000081539154},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.2874999940395355},{"id":"https://openalex.org/C117251300","wikidata":"https://www.wikidata.org/wiki/Q1849855","display_name":"Parametric statistics","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.2721000015735626},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.2709999978542328},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.2669999897480011},{"id":"https://openalex.org/C9158031","wikidata":"https://www.wikidata.org/wiki/Q6909140","display_name":"Moral reasoning","level":2,"score":0.2619999945163727}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3805713.3820397","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805713.3820397","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 International ACM SIGIR Conference on Innovative Concepts and Theories in Information Retrieval (ICTIR)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2604.17301","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2604.17301","pdf_url":"https://arxiv.org/pdf/2604.17301","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"doi:10.1145/3805713.3820397","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805713.3820397","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 International ACM SIGIR Conference on Innovative Concepts and Theories in Information Retrieval (ICTIR)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Detecting":[0],"harmful":[1],"content":[2],"in":[3,32,42,163,175],"multi":[4],"turn":[5,99,122],"dialogue":[6],"requires":[7,123],"reasoning":[8,50,101,126],"over":[9,152],"the":[10],"full":[11],"conversational":[12],"context":[13],"rather":[14],"than":[15],"isolated":[16],"utterances.":[17],"However,":[18],"most":[19],"existing":[20,130],"methods":[21],"rely":[22],"mainly":[23],"on":[24,133],"models":[25],"internal":[26],"parametric":[27],"knowledge,":[28],"without":[29,182],"explicit":[30,95],"grounding":[31],"external":[33,89],"normative":[34,96],"principles.":[35],"This":[36],"often":[37],"leads":[38],"to":[39],"inconsistent":[40],"judgments":[41],"socially":[43],"nuanced":[44],"contexts,":[45],"limited":[46],"interpretability,":[47],"and":[48,91,102,135,149,168],"redundant":[49,180],"across":[51,165],"turns.":[52],"To":[53,106],"address":[54],"this,":[55],"we":[56,109],"propose":[57],"RoTRAG,":[58],"a":[59,112,120],"retrieval":[60,124],"augmented":[61],"framework":[62],"that":[63,117,142],"incorporates":[64],"concise":[65],"human":[66],"written":[67],"moral":[68],"norms,":[69],"called":[70],"Rules":[71],"of":[72,160,173],"Thumb":[73],"(RoTs),":[74],"into":[75],"LLM":[76],"based":[77],"harm":[78,147],"assessment.":[79],"For":[80],"each":[81],"turn,":[82],"RoTRAG":[83,143],"retrieves":[84],"relevant":[85],"RoTs":[86],"from":[87],"an":[88,156,169],"corpus":[90],"uses":[92],"them":[93],"as":[94],"evidence":[97],"for":[98],"level":[100],"final":[103],"severity":[104,150],"classification.":[105],"improve":[107],"efficiency,":[108],"further":[110],"introduce":[111],"lightweight":[113],"binary":[114],"routing":[115],"classifier":[116],"decides":[118],"whether":[119],"new":[121],"grounded":[125],"or":[127],"can":[128],"reuse":[129],"context.":[131],"Experiments":[132],"ProsocialDialog":[134],"Safety":[136],"Reasoning":[137],"Multi":[138],"Turn":[139],"Dialogue":[140],"show":[141],"consistently":[144],"improves":[145],"both":[146],"classification":[148],"estimation":[151],"competitive":[153],"baselines,":[154],"with":[155],"average":[157,170],"relative":[158,171],"gain":[159],"around":[161],"40%":[162],"F1":[164],"benchmark":[166],"datasets":[167],"reduction":[172],"8.4%":[174],"distributional":[176],"error,":[177],"while":[178],"reducing":[179],"computation":[181],"sacrificing":[183],"performance.":[184]},"counts_by_year":[],"updated_date":"2026-07-16T05:54:29.318815","created_date":"2026-06-18T00:00:00"}