{"id":"https://openalex.org/W4411630262","doi":"https://doi.org/10.18653/v1/2024.eacl-long.1","title":"Enhancing Ethical Explanations of Large Language Models through Iterative Symbolic Refinement","display_name":"Enhancing Ethical Explanations of Large Language Models through Iterative Symbolic Refinement","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W4411630262","doi":"https://doi.org/10.18653/v1/2024.eacl-long.1"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2024.eacl-long.1","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.eacl-long.1","pdf_url":"https://aclanthology.org/2024.eacl-long.1.pdf","source":null,"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 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2024.eacl-long.1.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Xin Quan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xin Quan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032107191","display_name":"Marco Valentino","orcid":"https://orcid.org/0000-0002-9959-8385"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Marco Valentino","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040721287","display_name":"Louise A. Dennis","orcid":"https://orcid.org/0000-0003-1426-1896"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Louise Dennis","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5022138909","display_name":"Andr\u00e9 Victor Lucci Freitas","orcid":"https://orcid.org/0000-0002-5763-4990"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Andre Freitas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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":"1","last_page":"22"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9961000084877014,"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":0.9961000084877014,"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":0.986299991607666,"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/T13910","display_name":"Computational and Text Analysis Methods","score":0.9053999781608582,"subfield":{"id":"https://openalex.org/subfields/3300","display_name":"General Social Sciences"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6707239151000977},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.46950197219848633},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.3872215747833252},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3573276996612549}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6707239151000977},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.46950197219848633},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3872215747833252},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3573276996612549}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2024.eacl-long.1","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.eacl-long.1","pdf_url":"https://aclanthology.org/2024.eacl-long.1.pdf","source":null,"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 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2024.eacl-long.1","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.eacl-long.1","pdf_url":"https://aclanthology.org/2024.eacl-long.1.pdf","source":null,"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 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6661291015","display_name":"EnnCore: End-to-End Conceptual Guarding of Neural Architectures","funder_award_id":"EP/T026995/1","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"},{"id":"https://openalex.org/G8579816056","display_name":null,"funder_award_id":"200021_204617","funder_id":"https://openalex.org/F4320320924","funder_display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320319985","display_name":"Cancer Research UK","ror":"https://ror.org/054225q67"},{"id":"https://openalex.org/F4320319990","display_name":"National Institute for Health and Care Research","ror":"https://ror.org/0187kwz08"},{"id":"https://openalex.org/F4320320924","display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","ror":"https://ror.org/00yjd3n13"},{"id":"https://openalex.org/F4320333687","display_name":"Manchester Biomedical Research Centre","ror":null},{"id":"https://openalex.org/F4320334627","display_name":"Engineering and Physical Sciences Research Council","ror":"https://ror.org/0439y7842"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4411630262.pdf","grobid_xml":"https://content.openalex.org/works/W4411630262.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"An":[0],"increasing":[1],"amount":[2],"of":[3,16,69,156,175],"research":[4],"in":[5,37,47],"Natural":[6],"Language":[7,18],"Inference":[8],"(NLI)":[9],"focuses":[10],"on":[11,54,123],"the":[12,64,130,154,169],"application":[13],"and":[14,21,35,43,67,96,103,120,137,173],"evaluation":[15],"Large":[17],"Models":[19],"(LLMs)":[20],"their":[22,25,38,99],"reasoning":[23,145],"capabilities.Despite":[24],"success,":[26],"however,":[27],"LLMs":[28,84],"are":[29],"still":[30],"prone":[31],"to":[32,90,146,167],"factual":[33],"errors":[34],"inconsistencies":[36],"explanations,":[39],"offering":[40],"limited":[41],"control":[42],"interpretability":[44],"for":[45,159],"inference":[46],"complex":[48],"domains.In":[49],"this":[50],"paper,":[51],"we":[52,75],"focus":[53],"ethical":[55,70,125,141],"NLI,":[56],"investigating":[57],"how":[58],"hybrid":[59],"neurosymbolic":[60],"techniques":[61],"can":[62,112],"enhance":[63,168],"logical":[65,170],"validity":[66],"alignment":[68,174],"explanations":[71,95,114],"produced":[72],"by":[73],"LLMs.Specifically,":[74],"present":[76],"an":[77,86],"abductive-deductive":[78],"framework":[79],"named":[80],"Logic-Explainer,":[81],"which":[82],"integrates":[83],"with":[85],"external":[87],"backward-chaining":[88],"solver":[89],"refine":[91],"step-wise":[92],"natural":[93],"language":[94],"jointly":[97],"verify":[98],"correctness,":[100],"reduce":[101],"incompleteness":[102],"minimise":[104],"redundancy.An":[105],"extensive":[106],"empirical":[107],"analysis":[108],"demonstrates":[109],"that":[110],"Logic-Explainer":[111],"improve":[113],"generated":[115],"via":[116],"in-context":[117],"learning":[118],"methods":[119,158],"Chainof-Thought":[121],"(CoT)":[122],"challenging":[124],"NLI":[126,142,161],"tasks,":[127],"while,":[128],"at":[129],"same":[131],"time,":[132],"producing":[133],"formal":[134],"proofs":[135],"describing":[136],"supporting":[138],"models'":[139],"reasoning.As":[140],"requires":[143],"commonsense":[144],"identify":[147],"underlying":[148],"moral":[149],"violations,":[150],"our":[151],"results":[152],"suggest":[153],"effectiveness":[155],"neuro-symbolic":[157],"multi-step":[160],"more":[162],"broadly,":[163],"opening":[164],"new":[165],"opportunities":[166],"consistency,":[171],"reliability,":[172],"LLMs.":[176]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}