{"id":"https://openalex.org/W7171714348","doi":"https://doi.org/10.1145/3794763.3794807","title":"RlDecompiler: Enhancing LLM-based Decompilation via Reinforcement Learning with a Multi-Faceted Reward Function","display_name":"RlDecompiler: Enhancing LLM-based Decompilation via Reinforcement Learning with a Multi-Faceted Reward Function","publication_year":2026,"publication_date":"2026-04-12","ids":{"openalex":"https://openalex.org/W7171714348","doi":"https://doi.org/10.1145/3794763.3794807"},"language":null,"primary_location":{"id":"doi:10.1145/3794763.3794807","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3794763.3794807","pdf_url":null,"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 2026 34th IEEE/ACM International Conference on Program Comprehension","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3794763.3794807","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5143947919","display_name":"Yuchi Su","orcid":"https://orcid.org/0009-0000-1940-218X"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuchi Su","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0009-0000-1940-218X","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029715489","display_name":"Weina Niu","orcid":"https://orcid.org/0000-0002-3235-3463"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weina Niu","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0000-0002-3235-3463","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000589326","display_name":"Jiacheng Gong","orcid":"https://orcid.org/0009-0002-7986-2203"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiacheng Gong","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0009-0000-7970-9968","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071122921","display_name":"Ran Yan","orcid":"https://orcid.org/0009-0004-0728-3741"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ran Yan","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0009-0004-0728-3741","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100448208","display_name":"Song Li","orcid":"https://orcid.org/0000-0002-7961-8502"},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Song Li","raw_affiliation_strings":["Zhejiang University, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-7961-8502","affiliations":[{"raw_affiliation_string":"Zhejiang University, Hangzhou, China","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100352330","display_name":"Xin Liu","orcid":"https://orcid.org/0000-0003-3685-4852"},"institutions":[{"id":"https://openalex.org/I76214153","display_name":"Lanzhou University","ror":"https://ror.org/01mkqqe32","country_code":"CN","type":"education","lineage":["https://openalex.org/I76214153"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xin Liu","raw_affiliation_strings":["Lanzhou University, Lanzhou, China"],"raw_orcid":"https://orcid.org/0000-0003-3685-4852","affiliations":[{"raw_affiliation_string":"Lanzhou University, Lanzhou, China","institution_ids":["https://openalex.org/I76214153"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100780268","display_name":"Xiaosong Zhang","orcid":"https://orcid.org/0000-0001-9886-1412"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaosong Zhang","raw_affiliation_strings":["University of Electronic Science and Technology of China, Chengdu, China"],"raw_orcid":"https://orcid.org/0000-0001-9886-1412","affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"159","last_page":"170"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":null,"topics":[],"keywords":[{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.46889999508857727},{"id":"https://openalex.org/keywords/function","display_name":"Function (biology)","score":0.46869999170303345},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.2919999957084656},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.27559998631477356},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.27079999446868896}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5504999756813049},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.46889999508857727},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.46869999170303345},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46239998936653137},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2919999957084656},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.27559998631477356},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.25519999861717224},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.251800000667572},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3794763.3794807","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3794763.3794807","pdf_url":null,"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 2026 34th IEEE/ACM International Conference on Program Comprehension","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3794763.3794807","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3794763.3794807","pdf_url":null,"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 2026 34th IEEE/ACM International Conference on Program Comprehension","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":16,"referenced_works":["https://openalex.org/W2028509346","https://openalex.org/W2065489029","https://openalex.org/W2090554354","https://openalex.org/W2134312016","https://openalex.org/W2153150125","https://openalex.org/W2794718534","https://openalex.org/W3043641306","https://openalex.org/W4281387902","https://openalex.org/W4281808401","https://openalex.org/W4384948641","https://openalex.org/W4389208689","https://openalex.org/W4391725269","https://openalex.org/W4401752710","https://openalex.org/W4402671239","https://openalex.org/W4411522870","https://openalex.org/W7133243119"],"related_works":[],"abstract_inverted_index":{"Decompiling":[0],"binary":[1,127],"code":[2,7,186,215],"into":[3,128],"human-readable,":[4],"high-level":[5],"source":[6],"is":[8,93,188,216],"a":[9,32,56,94,144],"core":[10,88],"challenge":[11],"in":[12],"reverse":[13],"engineering.":[14],"While":[15],"traditional":[16],"methods":[17],"often":[18,44],"rely":[19],"on":[20,133],"brittle,":[21],"pattern-based":[22],"heuristics,":[23],"the":[24,53,64,68,101,126,167,194],"advent":[25],"of":[26,67,90,170,193,206],"Large":[27],"Language":[28],"Models":[29],"(LLMs)":[30],"offers":[31],"more":[33],"flexible":[34],"and":[35,60,122,156,173,183,200,208],"robust":[36],"approach.":[37],"However,":[38],"current":[39],"LLM-based":[40],"decompilation":[41],"efforts":[42],"are":[43],"limited":[45],"by":[46,143],"their":[47],"training":[48,81],"methodologies,":[49],"which":[50],"typically":[51],"treat":[52],"task":[54],"as":[55,116],"simple":[57],"sequence-to-sequence":[58],"translation":[59],"struggle":[61],"to":[62,83,110],"enforce":[63],"functional":[65,157],"correctness":[66,158],"output.":[69],"To":[70],"address":[71],"these":[72],"issues,":[73],"this":[74,134,163],"paper":[75],"proposes":[76],"an":[77,129],"innovative":[78],"framework":[79],"for":[80],"LLMs":[82],"perform":[84],"high-fidelity":[85],"decompilation.":[86],"A":[87],"contribution":[89],"our":[91],"work":[92],"novel":[95],"data":[96],"processing":[97],"pipeline":[98,105],"that":[99,148,178,192],"enriches":[100],"model\u2019s":[102],"input.":[103],"This":[104],"integrates":[106],"Ghidra-based":[107],"static":[108,117],"analysis":[109],"directly":[111],"embed":[112],"crucial":[113],"context,":[114],"such":[115],"resources":[118],"(strings,":[119],"floating-point":[120],"numbers)":[121],"relabeled":[123],"basic":[124],"blocks\u2014from":[125],"LLM-friendly":[130],"prompt.":[131],"Building":[132],"enriched":[135],"input,":[136],"we":[137,165],"employ":[138],"reinforcement":[139],"learning":[140],"fine-tuning":[141],"guided":[142],"multi-faceted":[145],"reward":[146],"function":[147],"comprehensively":[149],"evaluates":[150],"syntactic":[151],"correctness,":[152],"AST":[153],"similarity,":[154],"compilability,":[155],"via":[159],"test":[160],"cases.":[161],"Using":[162],"framework,":[164],"trained":[166],"RlDecompiler":[168,179,198],"family":[169],"models":[171,202],"(1.3B":[172],"3B).":[174],"Experimental":[175],"results":[176],"demonstrate":[177],"achieves":[180],"state-of-the-art":[181],"performance,":[182],"its":[184],"generated":[185],"quality":[187],"also":[189],"higher":[190],"than":[191],"baseline":[195],"models.":[196],"The":[197,214],"1.3B":[199],"3B":[201],"achieve":[203],"rerunnable":[204],"rates":[205],"27.96%":[207],"40.70%,":[209],"respectively,":[210],"outperforming":[211],"existing":[212],"baselines.":[213],"available":[217],"at":[218],"https://github.com/ri-char/rldecompile.":[219]},"counts_by_year":[],"updated_date":"2026-07-31T08:31:51.225901","created_date":"2026-07-30T00:00:00"}