{"id":"https://openalex.org/W7166815110","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1052","title":"Self-Correcting RAG: Enhancing Faithfulness via MMKP Context Selection and NLI-Guided MCTS","display_name":"Self-Correcting RAG: Enhancing Faithfulness via MMKP Context Selection and NLI-Guided MCTS","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166815110","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1052"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1052","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1052","pdf_url":"https://aclanthology.org/2026.findings-acl.1052.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":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.1052.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133597333","display_name":"Shijia Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shijia Xu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139819516","display_name":"Zhou Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou Wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139733489","display_name":"Xiaolong Jia","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaolong Jia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139793138","display_name":"Yu Guang Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139749264","display_name":"Kai Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kai Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133619831","display_name":"April Xiaowen Dong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"April Xiaowen Dong","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":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.80484626,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"20953","last_page":"20976"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.21770000457763672,"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":0.21770000457763672,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.21209999918937683,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.04439999908208847,"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/context","display_name":"Context (archaeology)","score":0.5789999961853027},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.516700029373169},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.25870001316070557},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.24650000035762787},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.23510000109672546}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5813999772071838},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5789999961853027},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.516700029373169},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.508400022983551},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.24650000035762787},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.23510000109672546},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.22370000183582306},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.21660000085830688},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2150000035762787}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1052","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1052","pdf_url":"https://aclanthology.org/2026.findings-acl.1052.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":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.1052","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1052","pdf_url":"https://aclanthology.org/2026.findings-acl.1052.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":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.4003952741622925,"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166815110.pdf","grobid_xml":"https://content.openalex.org/works/W7166815110.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Retrieval-augmented":[0],"generation":[1,43],"(RAG)":[2],"substantially":[3],"extends":[4],"the":[5,50,61,86,113],"knowledge":[6],"boundary":[7],"of":[8,115],"large":[9],"language":[10,93],"models.However,":[11],"it":[12],"still":[13],"faces":[14],"two":[15],"major":[16],"challenges":[17],"when":[18],"handling":[19],"complex":[20,134],"reasoning":[21,109,131],"tasks:":[22],"low":[23],"context":[24,65],"utilization":[25],"and":[26,42,47,78,111,121],"frequent":[27],"hallucinations.To":[28],"address":[29],"these":[30],"issues,":[31],"we":[32,53,89],"propose":[33],"Self-Correcting":[34],"RAG,":[35],"a":[36,68,82,91],"unified":[37],"framework":[38],"that":[39,126],"reformulates":[40],"retrieval":[41,58],"as":[44,67],"constrained":[45],"optimization":[46],"path":[48],"planning.On":[49],"input":[51],"side,":[52,88],"move":[54],"beyond":[55],"traditional":[56],"greedy":[57],"and,":[59],"for":[60],"first":[62],"time,":[63],"formalize":[64],"selection":[66],"multi-dimensional":[69],"multiplechoice":[70],"knapsack":[71],"problem":[72],"(MMKP),":[73],"thereby":[74],"maximizing":[75],"information":[76],"density":[77],"removing":[79],"redundancy":[80],"under":[81],"strict":[83],"token":[84],"budget.On":[85],"output":[87],"introduce":[90],"natural":[92],"inference":[94],"(NLI)-guided":[95],"Monte":[96],"Carlo":[97],"Tree":[98],"Search":[99],"(MCTS)":[100],"mechanism,":[101],"which":[102],"leverages":[103],"test-time":[104],"compute":[105],"to":[106],"dynamically":[107],"explore":[108],"trajectories":[110],"validate":[112],"faithfulness":[114],"generated":[116],"answers.Experiments":[117],"on":[118,133],"six":[119],"open-domain":[120],"multi-hop":[122],"QA":[123],"datasets":[124],"demonstrate":[125],"our":[127],"method":[128],"significantly":[129],"improves":[130],"accuracy":[132],"queries":[135],"while":[136],"effectively":[137],"reducing":[138],"hallucinations,":[139],"outperforming":[140],"strong":[141],"existing":[142],"baselines.":[143]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}