{"id":"https://openalex.org/W4415124804","doi":"https://doi.org/10.1109/icmlt65785.2025.11193242","title":"Enhanced Reverse Reasoning in LLMs for Further Research Directions Generation","display_name":"Enhanced Reverse Reasoning in LLMs for Further Research Directions Generation","publication_year":2025,"publication_date":"2025-05-23","ids":{"openalex":"https://openalex.org/W4415124804","doi":"https://doi.org/10.1109/icmlt65785.2025.11193242"},"language":"en","primary_location":{"id":"doi:10.1109/icmlt65785.2025.11193242","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icmlt65785.2025.11193242","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 10th International Conference on Machine Learning Technologies (ICMLT)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5092512988","display_name":"Meng Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210148850","display_name":"National Science Library","ror":"https://ror.org/04ndjyr10","country_code":"CN","type":"archive","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210148850"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Meng Wang","raw_affiliation_strings":["Chinese Academy of Science,National Science Library,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Science,National Science Library,Beijing,China","institution_ids":["https://openalex.org/I4210148850"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120312417","display_name":"Hui\u2010Jing Li","orcid":"https://orcid.org/0000-0002-2896-3867"},"institutions":[{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hanyu Li","raw_affiliation_strings":["University of Chinese Academy of Science,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Chinese Academy of Science,Beijing,China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000380008","display_name":"Jing Xie","orcid":"https://orcid.org/0000-0001-5624-458X"},"institutions":[{"id":"https://openalex.org/I4210148850","display_name":"National Science Library","ror":"https://ror.org/04ndjyr10","country_code":"CN","type":"archive","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210148850"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jing Xie","raw_affiliation_strings":["Chinese Academy of Science,National Science Library,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Science,National Science Library,Beijing,China","institution_ids":["https://openalex.org/I4210148850"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101648883","display_name":"Mengting Zhang","orcid":"https://orcid.org/0000-0002-7380-0397"},"institutions":[{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mengting Zhang","raw_affiliation_strings":["University of Chinese Academy of Science,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Chinese Academy of Science,Beijing,China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029389555","display_name":"Yajiao Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yajiao Wang","raw_affiliation_strings":["University of Chinese Academy of Science,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Chinese Academy of Science,Beijing,China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101598836","display_name":"Zhixiong Zhang","orcid":"https://orcid.org/0000-0003-1596-7487"},"institutions":[{"id":"https://openalex.org/I4210148850","display_name":"National Science Library","ror":"https://ror.org/04ndjyr10","country_code":"CN","type":"archive","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210148850"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhixiong Zhang","raw_affiliation_strings":["Chinese Academy of Science,National Science Library,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Science,National Science Library,Beijing,China","institution_ids":["https://openalex.org/I4210148850"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.13399518,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"355","last_page":"359"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10215","display_name":"Semantic Web and Ontologies","score":0.9743000268936157,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.9743000268936157,"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":0.9563000202178955,"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/construct","display_name":"Construct (python library)","score":0.5866000056266785},{"id":"https://openalex.org/keywords/masking","display_name":"Masking (illustration)","score":0.5662000179290771},{"id":"https://openalex.org/keywords/reverse-engineering","display_name":"Reverse engineering","score":0.5307000279426575},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5210999846458435},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.46860000491142273},{"id":"https://openalex.org/keywords/research-design","display_name":"Research design","score":0.4122999906539917},{"id":"https://openalex.org/keywords/experimental-research","display_name":"Experimental research","score":0.3968999981880188},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.37689998745918274}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6715999841690063},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.5866000056266785},{"id":"https://openalex.org/C2777402240","wikidata":"https://www.wikidata.org/wiki/Q6783436","display_name":"Masking (illustration)","level":2,"score":0.5662000179290771},{"id":"https://openalex.org/C207850805","wikidata":"https://www.wikidata.org/wiki/Q269608","display_name":"Reverse engineering","level":2,"score":0.5307000279426575},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5210999846458435},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.46860000491142273},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46320000290870667},{"id":"https://openalex.org/C2779318504","wikidata":"https://www.wikidata.org/wiki/Q1438035","display_name":"Research design","level":2,"score":0.4122999906539917},{"id":"https://openalex.org/C3018233652","wikidata":"https://www.wikidata.org/wiki/Q101965","display_name":"Experimental research","level":2,"score":0.3968999981880188},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.37689998745918274},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.3700999915599823},{"id":"https://openalex.org/C2986817661","wikidata":"https://www.wikidata.org/wiki/Q185698","display_name":"Research methodology","level":3,"score":0.357699990272522},{"id":"https://openalex.org/C2777626755","wikidata":"https://www.wikidata.org/wiki/Q28797","display_name":"Applied research","level":2,"score":0.34549999237060547},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3305000066757202},{"id":"https://openalex.org/C3018905339","wikidata":"https://www.wikidata.org/wiki/Q155207","display_name":"Case study research","level":2,"score":0.3172000050544739},{"id":"https://openalex.org/C2778734332","wikidata":"https://www.wikidata.org/wiki/Q7315082","display_name":"Research development","level":3,"score":0.3109000027179718},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.2971999943256378},{"id":"https://openalex.org/C2781287091","wikidata":"https://www.wikidata.org/wiki/Q947859","display_name":"Research proposal","level":2,"score":0.2955000102519989},{"id":"https://openalex.org/C539667460","wikidata":"https://www.wikidata.org/wiki/Q2414942","display_name":"Management science","level":1,"score":0.2858999967575073},{"id":"https://openalex.org/C8795937","wikidata":"https://www.wikidata.org/wiki/Q11862829","display_name":"Discipline","level":2,"score":0.2799000144004822},{"id":"https://openalex.org/C2986762682","wikidata":"https://www.wikidata.org/wiki/Q964754","display_name":"Basic research","level":2,"score":0.27480000257492065},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.26919999718666077}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icmlt65785.2025.11193242","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icmlt65785.2025.11193242","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 10th International Conference on Machine Learning Technologies (ICMLT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Further":[0],"research":[1,48,61,64,66,68,71,111,119,127,136,152,154,157,225,234,243,263,297],"direction":[2,264,298],"plays":[3],"a":[4,16,319],"pivotal":[5],"role":[6],"in":[7,39,103,110,118,126,135,184,224,233,242],"the":[8,36,55,83,93,97,141,189,200,213,261,267,272,276,289,309,313],"continuous":[9],"innovation":[10],"of":[11,219,258,283,311,328],"scientific":[12,51,329],"endeavors,":[13],"serving":[14],"as":[15,318],"critical":[17],"driving":[18],"force":[19],"for":[20,58,324],"disciplinary":[21],"advancement":[22],"and":[23,70,81,130,156,181,193,237,280,285,322],"knowledge":[24],"renewal.":[25],"This":[26],"study":[27],"proposes":[28],"an":[29,78,173,255],"enhanced":[30],"reverse":[31,163,191,227],"reasoning":[32,164,192],"method":[33,112,202,226,291],"based":[34],"on":[35,140],"masking":[37,205,273],"mechanism":[38,177,274],"large":[40],"language":[41],"models":[42,248],"(LLMs)":[43],"to":[44,92,160,178,295],"generate":[45],"high-quality":[46],"further":[47,60,262,296],"directions":[49],"from":[50],"literature.":[52],"We":[53],"analyze":[54],"meta-knowledge":[56],"required":[57],"generating":[59,325],"directions,":[62],"including":[63],"background,":[65,153],"questions,":[67,155],"methods,":[69,158],"limitations.":[72],"To":[73],"extract":[74],"meta-knowledge,":[75,143,312],"we":[76,144,171],"construct":[77],"instruction":[79],"dataset":[80],"fine-tune":[82],"LLM":[84],"using":[85],"QLoRA.":[86],"Experimental":[87,196],"results":[88,197],"show":[89],"that,":[90],"compared":[91],"baseline":[94],"model":[95,99,208,269,314],"Llama3.1-8B,":[96],"fine-tuned":[98,247,270],"achieves":[100],"significant":[101],"improvements":[102],"F1":[104,217,250],"scores:":[105],"+8.1%":[106],"(82.6%":[107],"vs.":[108,116,124,133],"90.7%)":[109],"extraction,":[113,121,129],"+8.2%":[114],"(82.0%":[115],"90.2%)":[117],"background":[120,244],"+8.5%":[122],"(81.4%":[123],"89.9%)":[125],"question":[128,235],"+8.9%":[131],"(80.7%":[132],"89.6%)":[134],"limitation":[137],"extraction.":[138],"Based":[139],"extracted":[142],"design":[145],"three":[146],"knowledge-driven":[147,204],"masked":[148],"prompt":[149],"templates":[150],"(masking":[151],"respectively)":[159],"train":[161],"LLMs\u2019":[162],"ability":[165],"by":[166,305],"occluding":[167],"key":[168],"information.":[169],"Additionally,":[170],"propose":[172],"adaptive":[174],"iterative":[175],"feedback":[176],"dynamically":[179],"identify":[180],"correct":[182],"biases":[183],"generated":[185],"content,":[186],"continuously":[187],"improving":[188],"model's":[190],"generalization":[194],"capabilities.":[195],"demonstrate":[198],"that":[199],"fine-tuning":[201],"with":[203,210,254,271],"significantly":[206],"enhances":[207],"performance,":[209],"ChatGLM3-13B":[211,268],"achieving":[212,278],"most":[214],"notable":[215],"results:":[216],"scores":[218,251,282],"91.7%":[220],"(a":[221,230,239],"6.3%":[222,231,240],"improvement)":[223,232,241],"reasoning,":[228,236],"90.7%":[229],"89.7%":[238],"reasoning.":[245],"All":[246],"maintain":[249],"above":[252],"87.0%,":[253],"average":[256],"improvement":[257],"6.0%.":[259],"In":[260],"generation":[265,299],"task,":[266],"performs":[275],"best,":[277],"ROUGE-L":[279],"BLEU-4":[281],"0.845":[284],"0.823,":[286],"respectively.":[287],"Overall,":[288],"proposed":[290],"can":[292,315],"be":[293],"applied":[294],"tasks":[300],"across":[301],"various":[302],"domains.":[303],"Furthermore,":[304],"updating":[306],"or":[307],"replacing":[308],"types":[310,327],"also":[316],"serve":[317],"theoretical":[320],"foundation":[321],"paradigm":[323],"other":[326],"texts.":[330]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-14T00:00:00"}